# Manage TextMine access and updates
Source: https://docs.textmine.com/admin-and-updates/index
Learn how to manage TextMine permissions, review current release notes, and find imported Intercom archive articles for older help content.
Use this section for feature access, permission questions, current product changes, and historical articles preserved from the Intercom help center.
## How do I manage TextMine permissions?
Start with [Permissions and feature access](/administration/permissions) to understand product access, API keys, MCP access, and admin checklist items.
## Where can I find product updates?
Use [Release notes](/release-notes) for current TextMine updates, including recent Workbench, workflow, Vault, folder, report, inbox, and Copilot Studio improvements.
## Where are older help center imports?
Use the [Intercom archive](/archive/intercom) for original imported articles that remain available for reference while current docs move into the help-center sections.
# Permissions and feature access
Source: https://docs.textmine.com/administration/permissions
Control access to newer TextMine surfaces.
TextMine exposes product areas through organization access, team membership, Vault scope, feature flags, guards, and permissions. Use this page as a practical guide for deciding who should be able to view, create, run, approve, publish, or export work.
## Permission principles
* Grant read access broadly only where the user should be able to see the underlying documents or data.
* Limit create/write access to users who can change reusable configuration or durable business data.
* Treat workflow triggers, provider imports, API keys, MCP tools, exports, and publication as higher-risk capabilities.
* Use approvals before actions that publish, email, submit externally, or update records at scale.
## Product access matrix
| Area | Read access allows | Create/write access allows | Higher-risk actions |
| ------------------- | ---------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- |
| Vault | View in-scope documents, folders, processing status, extracted fields, tables, images, and entities. | Upload/import documents, edit tags, assign reviewers, reprocess documents, validate values. | Bulk imports, reprocessing large scopes, changing document types, changing extraction settings. |
| Legislate | View published team documents. | Publish or manage validated documents where enabled. | Publishing from Vault to Legislate or treating unpublished Vault data as archive data. |
| Workbench | Run supervised analysis against permitted context. | Create outputs, import files into Workbench, request Vault import where enabled. | Importing provider files into Vault, running broad actions, using external tools. |
| Workflows | View workflow definitions and run history. | Create, edit, enable, run, or delete workflows depending on permission. | Sending emails, approvals, provider imports, record updates, external API calls, publishing to Legislate. |
| Playbooks | View and run available playbooks. | Create, edit, enable, disable, or delete playbooks. | Applying playbooks across many documents or using outputs for approvals/publication. |
| Records | View record schemas, records, candidates, evidence, and graph. | Create schemas, accept candidates, create/update/delete records. | Bulk record updates, schema changes, exports, enrichment that affects durable records. |
| Agents | View and run available agents. | Create or edit reusable agents. | Shared agent changes, scheduled/bulk runs, agent output used to mutate records or trigger workflows. |
| Scribe | View templates and generated documents. | Create, import, edit, publish, or fill templates. | Publishing templates, generating documents for external use, using sensitive source data. |
| Reports and exports | View generated reports and outputs. | Generate, download, email, export, or submit outputs. | External delivery, integration submission, large exports, reports without evidence. |
| Integrations | View connected providers and integration status. | Connect, configure, test, import from, or submit to providers. | Credential changes, provider disconnects, bulk imports, external submissions. |
| API and MCP | Read developer docs or discover tools. | Use API keys, bearer tokens, or MCP tools to act. | Secret creation, broad automated actions, destructive changes, high-volume data access. |
## Known permission labels
TextMine uses explicit permission labels for some newer surfaces. Current examples include:
| Area | Permission examples |
| --------- | -------------------------------------------------------- |
| Workflows | `workflow:read`, `workflow:create` |
| Playbooks | `playbook:read`, `playbook:create` |
| Records | `records:read`, `records:write`, `records:delete` |
| Vault | Controlled by Vault/team permissions and business scope. |
| Workbench | Guarded by feature access and user/team scope. |
Permission labels may be combined with feature flags and organization configuration. If a page or action is not visible, check both the user's permission set and whether the feature is enabled for the organization.
## API keys and MCP access
API and MCP access should be treated like operational access, not just documentation access.
Recommended rules:
* Create API credentials only for users or systems that need them.
* Store tokens in a secret manager or local environment variable, not in prompts or docs.
* Scope tokens to the minimum organization/team/data access available.
* Rotate credentials if they are pasted into chat, screenshots, issues, or logs.
* Confirm before using MCP/API tools that send emails, import provider files, update records, publish documents, or call external systems.
See [API and developer docs](/integrations/api-and-developer-docs) for the developer entry points.
## Admin checklist
Before enabling a user or team, decide:
1. Which Vaults, teams, and data scopes they should see.
2. Whether they can create reusable configuration such as workflows, playbooks, agents, schemas, or templates.
3. Whether they can trigger actions with side effects.
4. Whether they can approve or publish outputs.
5. Whether they can connect providers or manage credentials.
6. Whether they can export, email, or submit data externally.
7. Whether API/MCP access is needed and how credentials will be stored.
# Agents overview
Source: https://docs.textmine.com/agents/overview
Configure and run reusable AI agents.
Agents are reusable AI instructions that can run against documents, Vault context, Workbench context, workflow inputs, or other configured TextMine data. Use them when a repeated analysis needs consistent instructions but does not need the full branching, approval, and side-effect model of a workflow.
## When to use agents
| Use case | Use Agents when | Use something else when |
| ----------------------- | ----------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------- |
| Repeated analysis | The same prompt or reasoning task should run many times. | Use Workbench for one-off exploration. |
| Document review support | The agent should summarize, classify, compare, or extract reasoning from documents. | Use Playbooks for formal review policies and clause checklists. |
| Workflow step | The agent should produce an output inside a repeatable process. | Use a Workflow when the process itself has triggers, routing, approvals, emails, records, or integrations. |
| Product assistant | The agent should help a team follow a recurring review method. | Use Records when the output must become durable structured data. |
## Where agents run
Agents can be used from several surfaces:
* **Agents page**: create, edit, list, and run reusable agents where permitted.
* **Vault**: run an agent against selected documents or Vault context.
* **Workbench**: list agents, create a new agent when approved, run agents, and wait for completion.
* **Workflows**: use `run_agent` as a step in a repeatable automation.
* **API/MCP**: discover and run exposed agent operations where enabled.
## Agent inputs
An agent should state which context it expects. Common inputs include:
* Selected Vault documents.
* Uploaded Workbench files.
* Extracted tags, tables, entities, or records.
* A workflow run payload.
* A user-provided instruction or question.
* External context imported from a provider.
If an agent depends on processed Vault data, wait for document processing before running it. If it depends on tags, tables, or entities, confirm those upstream outputs are current.
## Agent outputs
Good agents produce predictable output formats. Examples:
| Output | Use it for |
| ----------------------- | ------------------------------------------------------------------- |
| Narrative summary | Human-readable analysis, review notes, or issue explanation. |
| Bullet list | Findings, risks, recommendations, or missing evidence. |
| Table | Comparisons, issue registers, review matrices, or extracted values. |
| JSON-like structure | Downstream workflow steps, API clients, or repeatable automation. |
| Decision with rationale | Escalation, approval support, or triage routing. |
For durable updates, connect the agent to a workflow, record update, report, or approval process. Do not rely on free-form output alone when the result changes business data.
## Run lifecycle
A typical agent run follows this lifecycle:
1. Select or discover the agent.
2. Confirm the scope, such as documents, Vault, record set, or workflow input.
3. Run the agent.
4. Wait for completion if the run is asynchronous.
5. Review the output and source evidence.
6. Use the output in Workbench, a workflow, a report, or a record process.
Agents should return run IDs and relevant document or record IDs when available so users can audit what happened.
## Good agent design
A strong agent definition includes:
* Purpose and boundaries.
* Required inputs.
* Output format.
* Source and evidence expectations.
* What the agent should not do.
* When to escalate to a human.
* Whether it can be used inside workflows.
Example instruction pattern:
```text theme={null}
Review the selected supplier agreement for renewal, termination, assignment, liability, and data protection risk. Return a table with issue, severity, source evidence, recommended action, and whether legal approval is required.
```
## Governance
Agents can have broad operational impact when run over many documents or embedded in workflows. Treat the following as higher-risk actions:
* Creating or editing shared agents.
* Running agents over large Vault scopes.
* Using agent output to update Records.
* Triggering workflows from agent output.
* Sending agent-generated reports externally.
* Scheduling agents to run without human review.
Agents should ask for approval before creating shared agents, changing reusable agent instructions, or running an agent across a large scope.
# How to create custom alerts
Source: https://docs.textmine.com/archive/intercom/document-lifecycle-management/how-to-create-custom-alerts
Set a range of alerts including expiry, notice, renewal and review notifications
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9227549-how-to-create-custom-alerts](https://help.textmine.com/en/articles/9227549-how-to-create-custom-alerts)
TextMine allows you to create custom alerts for your documents. These alerts can be for any of the following:
![Screenshot]()
Alerts can be sent to multiple people and you can specify a time and day for the alert which will be delivered by email.
## How to create an alert (Video)
To find out how to create an alert, what the following video or follow the step by step guide below.
## Step by step guide on how to create an alert
To create an alert, go to the Alerts tab in the top left of the side bar.
![Screenshot]()
You will see existing alerts which you can manage. You can also create new alerts with the create alert button.
The create alert button will ask you to specify the team followed by the document within that team for which you'd like to create an alert.
![Screenshot]()
![Screenshot]()
After specifying the document you can then specify the type of alert you'd like to receive followed by who should receive it and when.
Alerts can only be sent to users who have access to the document.
![Screenshot]()
Click on save once you are happy with the alert. It will now appear in your dashboard.
# How to sign a document
Source: https://docs.textmine.com/archive/intercom/document-lifecycle-management/how-to-sign-a-document
The steps to follow to sign a document in Legislate
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9227639-how-to-sign-a-document](https://help.textmine.com/en/articles/9227639-how-to-sign-a-document)
This article will explain how to sign a document in Legislate.
## When can you sign a document?
Documents can be signed once the terms have been set and parties on both sides have been added. The checklist tab will have a red tick next to the **Review and Sign** when the document is ready to be signed.
![Screenshot]()
You can either click on the **Review and Sign** button in the checklist or click on the **Sign** icon.
## Signing a document
To sign a document, click on the **Sign contract** button.
![Screenshot]()
You will then have the option to draw a signature or upload one.
![Screenshot]()
## Keeping track of signatures
Signatures are tracked on the signature tab.
![Screenshot]()
You can cancel a signature if the other side has not signed by clicking on the bin next to the user who has signed. The counter party will receive a notification when you sign.
![Screenshot]()
# How to use Scribe
Source: https://docs.textmine.com/archive/intercom/getting-started-with-textmine/how-to-use-scribe
Start creating smart templates
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9177917-how-to-use-scribe](https://help.textmine.com/en/articles/9177917-how-to-use-scribe)
Scribe is a template creation and management solution which allows you to define forms which can be used by the rest of the business to create instances of these documents quickly and safely!
![Screenshot]()
## Creating a new template
The first step is to create a template document during which you'll need to specify a folder where it will be stored and its type. You can create new folders and types or select from existing ones.
![Screenshot]()
## Drafting the template
Once the new template has been created, click on it to start drafting.
![Screenshot]()
You can format the text to your liking and insert variables, referred to as fields, by highlighting the relevant portion of text where they will be inserted.
![Screenshot]()
Doing this will create fields on the right. This fields form the basis of the form which users will need to complete in this example when issuing 1-Way NDAs from this template.
## Creating new Fields
When creating new fields which are specific to your document, select the portion of the text corresponding to the field and click on add new field. You'll then need to specify a Name, type, Label, validation and group.
A group allows you to organise your fields. The name is used for organising your fields internally whereas the label is the question which users will need to answer when completing the form.
![Screenshot]()
![Screenshot]()
## Managing fields
You can manage your fields at any point by clicking on the settings icon on the right hand side of the settings icon.
![Screenshot]()
## Comments and requests
Scribe has all the drafting functionality you'd expect such as requests, comments and flags. You can switch between editing and requesting from the top left.
![Screenshot]()
Edits will then be tracked and can be resolved by Scribe team mates.
![Screenshot]()
You can't approve your own requests.
## Publishing templates
Once you're ready to publish a template, you can preview the document and choose which teams should have access to it.
![Screenshot]()
![Screenshot]()
Please note that you can't publish a template with pending requests or comments.
Published templates can then be tracked in the Published tab of the Scribe dashboard.
By clicking on the 3 dots, you can republish a template to new teams or duplicate it if you need to make edits.
![Screenshot]()
## Creating instances of templates
Once you've published templates to Legislate teams, they can be configured by clicking on the create button. Select the document type you'd like to configure and its template version.
![Screenshot]()
You can then fill in the fields defined in Scribe to generate instances of your template.
![Screenshot]()
Once the document has been completed, you can invite parties for signature!
# How to use Vault
Source: https://docs.textmine.com/archive/intercom/getting-started-with-textmine/how-to-use-vault
Getting started with document data extraction
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9177886-how-to-use-vault](https://help.textmine.com/en/articles/9177886-how-to-use-vault)
Vault allows you to automatically extract key data points from business-critical documents using AI.
## Uploading documents
The first step is to upload a document by clicking on the upload button in the top right.
![Screenshot]()
You can drag and drop files or connect to your preferred file storage system. Click on Import once the documents have been uploaded.
Please note that Vault only supports PDF imports. However, Vault can work with PDF versions of scanned documents.
## Processing documents
Once the documents have been imported, Vault's AI will:
*
Classify the document type based on its substance
*
Extract key datapoints based on the document's detected type
![Screenshot]()
Vault will provide a confidence score to encourage you to verify the predictions. You can do this by clicking on the pen.
## Verifying predictions
After clicking on the pen, you'll see the uploaded pdf accompanied by the fields on the right which have been auto-populated by Vault's AI. The tags are either answering a question or extracting values from the document.
![Screenshot]()
Clicking on a field will take you to the relevant section of the document which was used to answer the question.
![Screenshot]()
Predictions can be rejected, corrected or validated.
Validate predictions one by one or in bulk by clicking on the **All button**.
You can then submit the document wit the submit button and the document will be pushed to Legislate where you'll be able to track the document amongst the rest of your document database!
# Welcome to TextMine!
Source: https://docs.textmine.com/archive/intercom/getting-started-with-textmine/welcome-to-textmine
An introduction to TextMine and its products
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9177855-welcome-to-textmine](https://help.textmine.com/en/articles/9177855-welcome-to-textmine)
Welcome to TextMine, the E2E knowledge base for your business critical documents! You can use TextMine to extract data from existing documents, explore the data and create new documents from your own template library.
TextMine has 3 components:
## Vault
Vault is a module for automatically tagging your business-critical documents with AI.
![Screenshot]()
## Legislate
Legislate is a module for exploring the data in your documents. Locate them one by one, answer aggregate queries and export raw data for further analysis.
![Screenshot]()
## Scribe
Scribe is a module for creating smart templates which can be used by the rest of the business to configure documents by completing forms.
![Screenshot]()
# Intercom archive
Source: https://docs.textmine.com/archive/intercom/index
Original imported Intercom articles preserved for reference.
# Intercom archive
These pages preserve the original Intercom help center content after conversion to MDX. Use the current product sections for maintained docs.
## Document Lifecycle Management
* [How to create custom alerts](/archive/intercom/document-lifecycle-management/how-to-create-custom-alerts)
* [How to sign a document](/archive/intercom/document-lifecycle-management/how-to-sign-a-document)
## Getting Started With Textmine
* [How to use Scribe](/archive/intercom/getting-started-with-textmine/how-to-use-scribe)
* [Welcome to TextMine!](/archive/intercom/getting-started-with-textmine/welcome-to-textmine)
* [How to use Vault](/archive/intercom/getting-started-with-textmine/how-to-use-vault)
## Legislate
* [A quick tour of Legislate (Video)](/archive/intercom/legislate/a-quick-tour-of-legislate-video)
* [Exploring document data with Advanced Search](/archive/intercom/legislate/exploring-document-data-with-advanced-search)
* [Creating contracts and documents in Legislate](/archive/intercom/legislate/creating-contracts-and-documents-in-legislate)
## Organisation Settings
* [User permission Settings](/archive/intercom/organisation-settings/user-permission-settings)
## Product Updates
* [26 May 2026](/archive/intercom/product-updates/26-may-2026)
* [8 May 2026](/archive/intercom/product-updates/8-may-2026)
* [25 March 2026](/archive/intercom/product-updates/25-march-2026)
* [12 March 2025](/archive/intercom/product-updates/12-march-2025)
* [5 March 2025](/archive/intercom/product-updates/5-march-2025)
* [19 February 2026](/archive/intercom/product-updates/19-february-2026)
* [12 February 2026](/archive/intercom/product-updates/12-february-2026)
* [28 January 2026](/archive/intercom/product-updates/28-january-2026)
* [19 January 2026](/archive/intercom/product-updates/19-january-2026)
* [17 December 2025](/archive/intercom/product-updates/17-december-2025)
* [5 December 2025](/archive/intercom/product-updates/5-december-2025)
* [26 Nov 2025](/archive/intercom/product-updates/26-nov-2025)
* [19 November 2025](/archive/intercom/product-updates/19-november-2025)
* [6 November 2025](/archive/intercom/product-updates/6-november-2025)
* [30 October 2025](/archive/intercom/product-updates/30-october-2025)
* [16 October 2025](/archive/intercom/product-updates/16-october-2025)
* [25 September 2025](/archive/intercom/product-updates/25-september-2025)
* [16 September 2025](/archive/intercom/product-updates/16-september-2025)
* [27 August 2025](/archive/intercom/product-updates/27-august-2025)
* [18 August 2025](/archive/intercom/product-updates/18-august-2025)
* [17 July 2025](/archive/intercom/product-updates/17-july-2025)
* [3 July 2025](/archive/intercom/product-updates/3-july-2025)
* [21 May 2025](/archive/intercom/product-updates/21-may-2025)
* [14 May 2025](/archive/intercom/product-updates/14-may-2025)
* [2 May 2025](/archive/intercom/product-updates/2-may-2025)
* [10 April 2025](/archive/intercom/product-updates/10-april-2025)
* [2 April 2025](/archive/intercom/product-updates/2-april-2025)
* [10 April 2024](/archive/intercom/product-updates/10-april-2024)
* [18 March 2025](/archive/intercom/product-updates/18-march-2025)
* [10 March 2025](/archive/intercom/product-updates/10-march-2025)
* [26 Feb 2025](/archive/intercom/product-updates/26-feb-2025)
* [14 Feb 2025](/archive/intercom/product-updates/14-feb-2025)
* [20 Feb 2025](/archive/intercom/product-updates/20-feb-2025)
* [6 February 2025](/archive/intercom/product-updates/6-february-2025)
* [30 January 2025](/archive/intercom/product-updates/30-january-2025)
* [22 January 2025](/archive/intercom/product-updates/22-january-2025)
* [18 November 2024](/archive/intercom/product-updates/18-november-2024)
* [December 2, 2024](/archive/intercom/product-updates/december-2-2024)
* [1 November 2024](/archive/intercom/product-updates/1-november-2024)
* [6 May 2024](/archive/intercom/product-updates/6-may-2024)
## Scribe
* [A quick tour of Scribe (Video)](/archive/intercom/scribe/a-quick-tour-of-scribe-video)
* [Template fields](/archive/intercom/scribe/template-fields)
* [How to add conditional logic into your Scribe templates](/archive/intercom/scribe/how-to-add-conditional-logic-into-your-scribe-templates)
* [Editing modes in Scribe](/archive/intercom/scribe/editing-modes-in-scribe)
* [How to import Word templates into Scribe](/archive/intercom/scribe/how-to-import-word-templates-into-scribe)
## Technical Faqs
* [How do I log out of TextMine?](/archive/intercom/technical-faqs/how-do-i-log-out-of-textmine)
## Vault
* [Vault supported documents and tags](/archive/intercom/vault/vault-supported-documents-and-tags)
* [Vault Utility Bill Tags](/archive/intercom/vault/vault-utility-bill-tags)
* [Vault Order Form Tags](/archive/intercom/vault/vault-order-form-tags)
* [Vault Purchase Order Tags](/archive/intercom/vault/vault-purchase-order-tags)
* [Vault Invoice Tags](/archive/intercom/vault/vault-invoice-tags)
* [Vault Service Agreement Tags](/archive/intercom/vault/vault-service-agreement-tags)
* [Vault NDA Tags](/archive/intercom/vault/vault-nda-tags)
* [Vault Consultancy Agreement Tags](/archive/intercom/vault/vault-consultancy-agreement-tags)
* [Vault Master Service Agreement Tags](/archive/intercom/vault/vault-master-service-agreement-tags)
* [Vault Supplier Agreement Tags](/archive/intercom/vault/vault-supplier-agreement-tags)
* [How to change a document type in Vault](/archive/intercom/vault/how-to-change-a-document-type-in-vault)
* [Creating manual fields in Vault](/archive/intercom/vault/creating-manual-fields-in-vault)
* [A quick tour of Vault (Video)](/archive/intercom/vault/a-quick-tour-of-vault-video)
* [Validating predictions in Vault](/archive/intercom/vault/validating-predictions-in-vault)
* [An introduction to Vault teams](/archive/intercom/vault/an-introduction-to-vault-teams)
* [Making corrections in Vault](/archive/intercom/vault/making-corrections-in-vault)
# A quick tour of Legislate (Video)
Source: https://docs.textmine.com/archive/intercom/legislate/a-quick-tour-of-legislate-video
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9227437-a-quick-tour-of-legislate-video](https://help.textmine.com/en/articles/9227437-a-quick-tour-of-legislate-video)
In this video, we provide a quick tour of Legislate's document creation and data exploration capabilities.
# Creating contracts and documents in Legislate
Source: https://docs.textmine.com/archive/intercom/legislate/creating-contracts-and-documents-in-legislate
How to create instances of Scribe templates in Legislate
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9220892-creating-contracts-and-documents-in-legislate](https://help.textmine.com/en/articles/9220892-creating-contracts-and-documents-in-legislate)
Legislate allows users to create instances of Scribe templates via forms. For example, the legal team might have an employment contract template which the HR team will typically fill in and send to new hires. With Legislate, the HR team will only need to fill in a form to generate the final version of the employment contract template for the new hires.
This article shows you how to create documents in Legislate from Scribe templates!
## Publishing Scribe templates to Legislate teams
Scribe allows you to choose which teams in Legislate will have access to any given Scribe template.
![Screenshot]()
Legislate users do not have access to Scribe by default which means that a Scribe editor will need to provide template access to Legislate users.
## Creating new documents from Scribe templates
Once a template has been published to the correct teams, it can be created within a Legislate team.
When in the team, click on the Create button and provide a name of the folder where this document will be stored. You can also create a new document within an existing folder.
![Screenshot]()
You can then select the Scribe template version and click on **Continue**.
![Screenshot]()
You can then choose to copy terms over from a previous document or start from scratch.
![Screenshot]()
Once this has been done, you'll then be presented with the Scribe form which will need to be completed in order to render the complete document.
![Screenshot]()
After filling in the form and clicking on Save, you'll be able to visualise the document in the preview tab.
![Screenshot]()
You can add parties to the document via the people tab. A party can be a company or an individual. When specifying the company, you will be required to specify the signatory.
![Screenshot]()
You can also provide access to document collaborators as well as witnesses and supporting signatures via the **Other people** button.
# Exploring document data with Advanced Search
Source: https://docs.textmine.com/archive/intercom/legislate/exploring-document-data-with-advanced-search
Quickly locate and export document data in Legislate
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9220936-exploring-document-data-with-advanced-search](https://help.textmine.com/en/articles/9220936-exploring-document-data-with-advanced-search)
Legislate's Advanced search functionality allows you to quickly locate documents based on dynamic questions and aggregate queries.
This article will show you how!
## Advanced Search
Advanced search is located below the search bar and allows you to quickly filter your documents based on the data they contain. The filters are generated dynamically based on the underlying data in your documents which means that you'll only see filters which exist.
![Screenshot]()
When selecting a filter, the list of documents and the other filters will update in real time.
![Screenshot]()
![Screenshot]()
Once you have selected the filters you are looking to apply, you can see the documents which match the criteria and export the data as a CSV.
![Screenshot]()
The exported CSV will then contain the documents matching the search criteria and the corresponding fields.
![Screenshot]()
This process automated document audits which means that you can easily report on your documents to stakeholders without being distracted from your important day-to-day duties!
# User permission Settings
Source: https://docs.textmine.com/archive/intercom/organisation-settings/user-permission-settings
This article describes the User permissions which can be set by an Admin of an Organisation
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11364357-user-permission-settings](https://help.textmine.com/en/articles/11364357-user-permission-settings)
The admin of an organisation only has access to the Admin portal which can be used to create business units, roles and assign members of the organisation to a role and/or business units.
![Screenshot]()
## Permissions
The following permissions, organised by TextMine section can be assigned to a role:
**Legislate**
Permission
Description
Access to Legislate and see contracts/documents
Archive and activate a contract/document
Archive/Unarchive documents
Create a contract
Only applicable to Scribe customers
Delete a contract/document
Download CSV and PDF
Move a document to Vault
Update folder name
Update name of a contract/document
### **Vault**
Permission
Description
Access to Vault and see Documents
Assign, change and remove assignee
Assign, change and remove reviewer
Change type and reprocess documents and tags
Create a custom tag for a document
Manually insert a tag in the plain text
Highlight values in the plain text tab and assign them to a default tag.
![Screenshot]()
Delete a custom tag for a document
![Screenshot]()
Delete a manually created tag or a tag with no answer
Delete a manually created tag or a tag with no answer
![Screenshot]()
![Screenshot]()
Delete documents
Download documents in any format
Download documents from Vault as a marked up PDF or plain text
![Screenshot]()
Edit document name
Edit tag values within a document
Make corrections to the values of a tag
![Screenshot]()
Process custom tags
Submit documents to Legislate
Update a custom tag prompt for a document
![Screenshot]()
Upload new documents
Validate or reject a tag within a document
![Screenshot]()
Create personal team **(As of 21 May 2025)**
Allows users to have a draft space to experiment with Vault. A personal team is not a Business Unit.
![Screenshot]()
**Vault Settings**
Permission
Description
Create a custom tag in Vault settings
Create a custom tag which will apply to all documents uploaded to the Vault
![Screenshot]()
Delete custom tags in Vault settings
Edit the default settings of a Vault Business Unit
Enable/disable signature detection, agile text extraction and the other Vault configuration settings of a Business Unit.
![Screenshot]()
Enable and disable custom tags in Vault settings
Create custom document types and assign them to a business unit
Create custom document types and associated tags.
![Screenshot]()
Enable users to select which custom prompts apply to which document types
Enable users to promote a custom tag to a specific document as a default tag
### Other
Create and manage Alerts
Create and manage Agents
The permissions assigned to a role can be updated by the Organisation Admin at anytime.
# 1 November 2024
Source: https://docs.textmine.com/archive/intercom/product-updates/1-november-2024
Custom flows available
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10073053-1-november-2024](https://help.textmine.com/en/articles/10073053-1-november-2024)
TextMine now allows users to skip the TextMine default document processing pipeline in Vault. When uploading a new document to Vault, simply untick the "Keep default TextMine field" check box.
![Screenshot]()
In this setting, TextMine will run the enabled vault team tags against the document.
![Screenshot]()
Once the document has been processed, the type will be listed as unspecified and you click on it to view the extracted information:
![Screenshot]()
![Screenshot]()
Please note that by deselecting the TextMine fields option, the model will not currently classify the document or look to identify signatures.
# 10 April 2024
Source: https://docs.textmine.com/archive/intercom/product-updates/10-april-2024
Bug fixes and Scribe release
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9229963-10-april-2024](https://help.textmine.com/en/articles/9229963-10-april-2024)
TextMine has released the following bug fixes:
*
Bugs related to the search bar
*
Bugs related to the file storage integrations
TextMine has released Scribe from beta to general availability. The main aspect of this release is that Scribe can now support custom templates and [fields](https://help.textmine.com/en/articles/9220737-template-fields).
The following video provides a demo of Scribe and visit our [Scribe collection](https://help.textmine.com/en/collections/9145553-scribe) for tutorials.
# 10 April 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/10-april-2025
Custom document types
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11064676-10-april-2025](https://help.textmine.com/en/articles/11064676-10-april-2025)
When uploading document to Vault, the model will classify documents based on one 60 types and then extract the key terms based on the type. Previously, users would need to submit requests to TextMine to add new types.
It is now possible for users to create their own custom document types and associated tags in Vault.
## How to create a custom document type
Go to the organisation settings under the Vault section in the sidebar and create a new document type.
![Screenshot]()
You will then have to specify a name and category for the new document type as well as which Vault teams should have access to the custom document type.
![Screenshot]()
Avoid acronyms to improve the accuracy of the classification. For example, call a type Non-Disclosure Agreement instead of NDA. However, it can make sense to name a new type based on the acronym if the acronym or code is how the document type is referred to e.g. Form ADV, 10-Q, etc.
Once you've created your document type you will see the document type in the list. Click on the type to create tags for the new custom type.
![Screenshot]()
![Screenshot]()
You can enable/disable the tags for the type.
![Screenshot]()
## Predicting custom document types
Once a custom document type and associated tags has been enabled for a team, you can activate the custom document type pipeline from the team's Vault settings in the top right.
![Screenshot]()
Then enable custom document types.
![Screenshot]()
If this setting is enabled, Vault will then classify the uploaded documents amongst the custom types you have created.
![Screenshot]()
![Screenshot]()
You can also update the type of a document to a custom type by clicking on the type field of the document and selecting the corresponding type from the list.
![Screenshot]()
The list of types will correspond to your custom types only. To change to a TextMine default, please read the following section.
## Changing a default document type to a custom type and vis versa
To change a custom type to a default type, disable custom types in the team settings.
![Screenshot]()
When you select the type in the team view, you will then be able to see the list of default types.
![Screenshot]()
If you would like to change a default type to a custom type, select custom document types from the settings. When you click on the type the list will be limited to the custom types.
Whilst we are pro-active with fine-tuning the models, we might not have done training for new custom types and tags. Let us know if you are seeing inconsistent performance and we can improveme the models with additional fine-tuning.
## Video Walkthrough
# 10 March 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/10-march-2025
Multiple reviewers, customise columns, new home screen
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10742609-10-march-2025](https://help.textmine.com/en/articles/10742609-10-march-2025)
We have released a number of key improvements to the platform:
## Multiple reviewers and better reviewer management
It is now possible to have multiple reviewers. An assignee can do the first analysis of Vault tags and a reviewer can double check the work of the assignee.
![Screenshot]()
You can filter the documents in a Vault team by document type, uploader or assignee.
![Screenshot]()
You can also filter the tags by Reviewer in a Vault document:
![Screenshot]()
## New home screen for Vault users
Vault users can now see recently viewed documents and the documents which have been assigned to them.
![Screenshot]()
## Customise Vault columns
You can configure which columns to display in a Vault team by clicking on the Columns icon below Automated insights. The settings will be saved for the next time you visit the team.
![Screenshot]()
## Agile text extraction
This setting allows you to prioritise quick text extraction techniques over more expensive extraction techniques which might take longer. This can be enabled in the Vault settings:
![Screenshot]()
# 12 February 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/12-february-2026
Alternative names for document types, on demand agents
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13709723-12-february-2026](https://help.textmine.com/en/articles/13709723-12-february-2026)
## Alternative names
It is now possible to provide alternative names for a document type which can be used to classify the document. This is useful when the document type name is an abbreviation or not descriptive. Alternative names also increase the chances of the model correctly classifying the document as the desired type.
![Screenshot]()
## Chained default tags for images
You can now create default chained tags for images which is useful for running prompts on image extracts systematically when the image type is detected.
![Screenshot]()
## On-demand agents
You can now create on-demand agents which can be run from Vault. On-demand also offer more flexibility as you don't need to pre-define the scope of the analysis.
![Screenshot]()
Once you have created your on-demand agent, specify the instruction you would like it to run.
![Screenshot]()
Then within the Vault, select the documents you'd like to run the agent on and click on Run agent.
![Screenshot]()
You'll then be able to select which agent you'd like to use and see the processing in the sidebar.
![Screenshot]()
The results are then accessible from the Agents section or by clicking on the completed run in Vault. The results from consecutive runs will append in the on-demand agent's results summary.
![Screenshot]()
## Table improvements
Extracted tables can be included in Agent reports. This allows the reports to be leverage more data than the tags. You can now also extract tables manually from image files.
## Folder/Zip import
You can now import a zip or folders into Vault. Vault will recreate the folder structure based on what was uploaded.
![Screenshot]()
## Email integration
You can email attachments to Vault directly. Each Vault has its own email and emails can be refreshed or disabled on demand. You can choose to include the email body in the analysis. When enabled, Vault will be able to answer questions about the contents of the email (e.g. instructions).
![Screenshot]()
## Ariba integration
The Ariba integration allows you to connect your Ariba documents to Vault and make the reverse export easier with intelligent Ariba supplier matching.
### Import contract workspace and project documents
![Screenshot]()
### Match supplier details to Ariba
When the Supplier Data API is enabled, Vault will match supplier names identified in documents to their corresponding Ariba IDs.
![Screenshot]()
# 12 March 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/12-march-2025
Tabular views and Open Corporates API
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/14046582-12-march-2025](https://help.textmine.com/en/articles/14046582-12-march-2025)
## Tabular Views
Data in Vault can now be view from the Vault team in tabular format. To activate this, click on the table icon.
![Screenshot]()
You can choose which columns to include and adjust additional settings for each column.
![Screenshot]()
![Screenshot]()
## Open Corporates
You can now access open corporate data in the platform with your Open Corporates API key.
![Screenshot]()
# 14 Feb 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/14-feb-2025
Improved dashboard, tag highlighting and reviewing
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10559277-14-feb-2025](https://help.textmine.com/en/articles/10559277-14-feb-2025)
## Vault Dashboard
The Vault dashboard now provides greater visibility into document uploads. This now includes the upload date, uploader name, reviewer name and review date. All columns can be used to sort the documents in Vault.
![Screenshot]()
## Tag highlighting
Within a document, it is now possible to see the highlighted tags in the pdf preview. This was previously only possible in the plain text. A key thing to note is that the highlighting might not work on scanned documents. However, the search will always work on the Plain text tab.
![Screenshot]()
You can also use the search bar to navigate through the document. Using your keyboard's arrows you can switch through the answers.
![Screenshot]()
## Reviewers
### Assigning Reviewers
You can now assign uploaded documents to reviewers.
#### Assign reviewers in bulk:
Select the relevant documents and assign the reviewer
![Screenshot]()
![Screenshot]()
#### Assign reviewers one by one
Click on the reviewer label and assign a reviewer. You can also un-assign or re-assign a document's reviewer at any time.
![Screenshot]()
![Screenshot]()
### Audit Trail
The review date now appears next to individual tags when a field is validated.
![Screenshot]()
The document's overall review date now appears in the dashboard.
![Screenshot]()
### Validation settings
Within the Vault settings it is now possible to enforce the review strategy by enabling the "Only reviewers can validate fields" setting.
![Screenshot]()
This means that unassigned tags can't be reviewed by users who are not assigned to the document.
# 14 May 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/14-may-2025
Breadcrumbs
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11393387-14-may-2025](https://help.textmine.com/en/articles/11393387-14-may-2025)
We have improved the navigation experience by introducing breadcrumbs at the top of the page.
![Screenshot]()
Breadcrumbs helps users quickly assess where they are in the application and easily navigate to their desired teams or subfolders by clicking on the breadcrumb heading.
![Screenshot]()
# 16 October 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/16-october-2025
Folders in Vault
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12595058-16-october-2025](https://help.textmine.com/en/articles/12595058-16-october-2025)
When importing documents into Vault, you can now organise them as folders. Select an existing folder or create a new one.
![Screenshot]()
After the documents have been imported into a folder, you can add the folder column to the view and filter the Vault by that folder.
![Screenshot]()
![Screenshot]()
You can also move documents between folders by selecting them and clicking on the move to folder icon. You will then be prompted to confirm which folder you would like to send the documents to.
![Screenshot]()
You can view all the folders of the Vault under the Folders tab.
![Screenshot]()
You can then click on a folder to see all the documents inside that folder and you can edit the name of the folder if required.
![Screenshot]()
When you submit documents to Legislate, the folders will be carried through.
# 16 September 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/16-september-2025
Agent Actions
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12302368-16-september-2025](https://help.textmine.com/en/articles/12302368-16-september-2025)
Agents can now perform actions based on Vault results. The first action is based on the confidence score of a document and allows you to assign or submit it based on the score. To get started go to an Agent with an existing source and select Add Action.
![Screenshot]()
You can determine whether the agent should run on existing document or only new documents uploaded to the connected Vault. You can configure the conditions and actions to your requirements.
![Screenshot]()
You will see a summary of the results once the agent has finished running.
![Screenshot]()
We are expanding the list of actions including the ability to apply rules against the values extracted by Vault. Get in touch if you would like to discuss your use case and how it can be solved by TextMine's agents!
# 17 December 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/17-december-2025
Reasoning tags, Agent exports and Team improvements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13177269-17-december-2025](https://help.textmine.com/en/articles/13177269-17-december-2025)
## Reasoning Tags
We now offer reasoning tags for both custom tags and default document type tags. They can also be applied to aggregation tags.
![Screenshot]()
When enabled, the model will provide additional reasoning to how it found the answer. You can choose to include it in the answer.
![Screenshot]()
![Screenshot]()
## Agent Results and exports
Agent results now present links to the source files and expandable cells to better interpret the results. The results table can also be exported to CSV.
![Screenshot]()
![Screenshot]()
## Improved Legislate Team view
The Legislate team view now offers configurable columns and allows you to also view when documents were published from Vault to Legislate as well as the assignee and export date.
![Screenshot]()
![Screenshot]()
## Delete personal working teams
You can now delete working (personal) teams as long as they don't have any documents in either Vault or Legislate. Business Units/teams connected to an organisation can't be deleted.
![Screenshot]()
![Screenshot]()
# 17 July 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/17-july-2025
Document Type and Agent improvements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11792925-17-july-2025](https://help.textmine.com/en/articles/11792925-17-july-2025)
## Document Type Improvements
It is now possible to choose which Document Types, including types created by TextMine, should be included in the document classification.
![Screenshot]()
Your active document types are present in the Document Types tab and you can enable/disable them for a specific Vault at any time.
You can click on the document type to view the associated default tags.
![Screenshot]()
The default tags can be enabled or disabled as required.
Document types in the TextMine library can be promoted to your Document Types by selecting the types you wish to promote and clicking on promote.
![Screenshot]()
Once a Document Type is associated with a Vault, it will be used as part of the classification process. As a result, you no longer need to make special configurations in the Vault settings.
![Screenshot]()
## Agent improvements
Agents can now pull 10-Ks, 20-Fs for companies present in a CSV, xls or xslx.
![Screenshot]()
Download the example CSV and complete it with company names.
![Screenshot]()
Review the suggestions and save the agent.
![Screenshot]()
This will then save the companies and submit the latest 10Ks, 20Fs for those companies to the corresponding Vault.
![Screenshot]()
# 18 August 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/18-august-2025
Analytics
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12037212-18-august-2025](https://help.textmine.com/en/articles/12037212-18-august-2025)
Organisations can now track high level analytics across their business units. These analytics include the status of documents uploaded to Vault, the overall accuracy by document type and the number of pages processed by their team.
![Screenshot]()
The tracking period for each chart can be changed and your preference will be saved automatically based on your select.
![Screenshot]()
To enable analytics for a specific role, the admin of the organisation will need to enable the "View Vault analytics" permission under Vault permissions.
![Screenshot]()
![Screenshot]()
# 18 March 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/18-march-2025
Support for ppt, xlsx and performance improvements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10816997-18-march-2025](https://help.textmine.com/en/articles/10816997-18-march-2025)
## Powerpoint and excel files can now be imported into Vault
Vault now supports powerpoint and excel files.
![Screenshot]()
The full list of file types supported by Vault is:
*
PDF
*
Word
*
Excel
*
Powerpoint
Vault can also work with scans and images as long as they are embedded into these file types.
## Tag highlighting
We have fixed some edge cases where the tag highlighting wasn't working in the pdf preview and have improved the ordering.
![Screenshot]()
## Download the plain text of a document
You can now download the extracted text of a document in whole when clicking on the download button of the plain text tab.
![Screenshot]()
# 18 November 2024
Source: https://docs.textmine.com/archive/intercom/product-updates/18-november-2024
Pagination and pdf exports
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10191547-18-november-2024](https://help.textmine.com/en/articles/10191547-18-november-2024)
TextMine has released a number of new features to improve the user experience. These include:
*
Pagination of documents in Vault
*
The ability to jump to a relevant page on the preview as opposed to the plain text
*
The ability to generate a marked up pdf containing the
## Pagination
Users can go to a specific page or select a field to go to the page which contains the field. Users can also switch between the preview and plain text of a specific page.
![Screenshot]()
## Marked up exports
You can now generate a marked up copy of the preview which can be easily downloaded and shared with relevant stakeholders.
![Screenshot]()
The pdf can be generated with the download button next to the plain text toggle. The fields will be listed on the corresponding page where the model found the answer.
# 19 February 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/19-february-2026
Embedded image detection, org chart transcription, marked up image downloads
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13763619-19-february-2026](https://help.textmine.com/en/articles/13763619-19-february-2026)
## Embedded image detection
It is now possible to identify images inside documents. Images are classified against your predefined image types and their corresponding tags will be processed against any detected images.
For example, if you have a chart image type and your default tags ask for a summary of the chart, the embedded image detection will provide a summary for each chart detected in your document.
To get started, define your image types, tags and associated vaults.
![Screenshot]()
You can then either enable embedded image detection at the document level by going to the document type's settings.
![Screenshot]()
Alternatively, you can enable embedded image detection at the Vault level in which case it will run on all documents uploaded to the Vault.
![Screenshot]()
You can then view the detected images and associated tags in the documents in Vault.
![Screenshot]()
## Org chart transcription
When processing a "Chart Transcription" tag in Org charts, Vault will follow a specific pipeline in order to accurately transcribe the entities and relationships. Entities are sorted by order of appearance for ease of review and corrections.
![Screenshot]()
## Marked up image downloads
Image documents can now be exported to marked up PDFs with their tags.
![Screenshot]()
# 19 January 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/19-january-2026
Image types, table extraction, and more
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13431363-19-january-2026](https://help.textmine.com/en/articles/13431363-19-january-2026)
## Image Types
You can now have document types for images and Vault now supports image uploads (jpeg, png). To do this, create a document type and select image file type. You can associate the image type to the relevant Vaults.
![Screenshot]()
You can then create default tags for the image.
![Screenshot]()
Once the image tags are created, you can upload the image to the corresponding Vault.
![Screenshot]()
You can then review the image and associated tags.
![Screenshot]()
## Table extraction
TextMine now offers table extraction from documents. You can either extract tables via a manual selection or enable it automatically for the Vault.
### Manual selection
Click on extract table from Image and use the cursor to select the table in the preview.
![Screenshot]()
![Screenshot]()
Once the extraction is complete, you'll see the table in the Table tab. You can edit the values or download the extracted table to CSV.
![Screenshot]()
To cross-reference the results more easily, click on the pop up arrow above CSV to pop out the table in new window.
![Screenshot]()
To enable automated table extraction for the Vault, go to the Vault settings and enable it in the Default settings. This will apply to all documents uploaded to the Vault.
![Screenshot]()
## Chained tags in Document Types
It is now possible to create default chained tags.
![Screenshot]()
## Pop out tags
You can now pop out the tags cards to a new window in order to simplify the document/tag cross-referencing.
![Screenshot]()
The same functionality as in the editor is mirrored in the pop out.
![Screenshot]()
## Zoom in and out
You can now change the zoom resolution of the document upload.
![Screenshot]()
## Analytics improvements
The Analytics are now tracking additional metrics such as the most edited tags, errors, time spent reviewing and general user activity.
![Screenshot]()
The Documents and Error charts also allow you to assign documents to users if required.
![Screenshot]()
## Summary improvements
The summary functionality has been improved to offer more accurate and meaningful results.
![Screenshot]()
## Edit Agent Actions
You can now edit agent actions. Changing the action or instruction will reprocess the agent.
![Screenshot]()
# 19 November 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/19-november-2025
Image tags, Chained tags, List and folder improvements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12870171-19-november-2025](https://help.textmine.com/en/articles/12870171-19-november-2025)
## Folder improvements
We now display folders in the sidebar below the business unit.
![Screenshot]()
In addition to being able to filter by folder, you can now sort the Vault view by Folder.
![Screenshot]()
## Vault editor improvements
The vault editor performance and usability has been improved.
When validating tags, you will be taken to the next tag to be validated automatically.
When a tag presents data in a list format, we will suggest a list editing tool to make the handling of the data easier.
![Screenshot]()
You can search, keep or delete the selection.
## Image Tags
You can now prompt images within documents. Click on create image tag and select the portion of the image you would like the model to analyse.
![Screenshot]()
Once you've selected the image, create and process your question.
![Screenshot]()
You will then see the answers as image tags the editor.
![Screenshot]()
## Chained Tags
A chained tag is a new tag which allows you to answer questions using answers from other tags. To created a chained tag, click on custom tags and enabled chained tag.
![Screenshot]()
This will open a pop up with tags you can use as source tags to your question.
![Screenshot]()
Once you've selected the source tags, create your question.
![Screenshot]()
Once you run the chained tag, the answer will display with a chain icon.
![Screenshot]()
You can run chained tags on any type of tag including previously run chained tags.
## Company search improvements
We now apply fuzzy logic to company name searches to overcome typos, punctuation or abbreviations.
![Screenshot]()
![Screenshot]()
## Timezones
The times in the app are now displayed based on the user's browser. However, times in PDF exports are still based on the server's timezone.
## General Actions
General agent actions can now be applied to folders. This allows the agent to answer questions looking at the extracted data from all documents in a folder.
![Screenshot]()
# 2 April 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/2-april-2025
Bulk delete and pagination
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11017269-2-april-2025](https://help.textmine.com/en/articles/11017269-2-april-2025)
It's now possible to delete documents in Vault in bulk. Simply select the document you wish to delete and press the delete button icon.
![Screenshot]()
We have also introduced pagination which will accelerate the page loading times for teams with a significant amount of uploads. Pagination preferences are saved per user and team.
# 2 May 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/2-may-2025
Admin portal and access controls
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11183590-2-may-2025](https://help.textmine.com/en/articles/11183590-2-may-2025)
## Organisations
We have released a new admin portal which allows enterprise administrators to configure granular user access controls based on specific roles.
Once TextMine has created an organisation for your company, an admin user will be created and shared with your company's IT administrator. The IT administrator will then be able to define the roles and business units of the organisation. Members of your company can then be added to the organisation in TextMine via an active directory.
## Members
An admin can view the members of the organisation as well their roles and which business units they are part of. Members can be added to an organisation via an active directory.
![Screenshot]()
## Business Units
An admin can create as many business units as required. A business unit is the equivalent to a team in Vault/Legislate.
![Screenshot]()
By clicking on a business unit, the admin can add or remove members from the business unit.
![Screenshot]()
For organisation users, this approach replaces the previous approach of adding users to a team via Vault.
## Roles
Admins can create as many roles as required depending on the needs and permissions of their organisation. For each role, the admin can configure the permitted read and write operations of that role throughout the TextMine application.
![Screenshot]()
The permissions of a role can be updated by the admin based on the requirements of the organisation.
## Access roles in practice
Most of the user experience will remain the same for organisation members. However, when logged in, an organisation member will only see the business units they have access to. Functionality within the application will be enabled or disabled depending on their role.
![Screenshot]()
![Screenshot]()
If required, organisation members can still have the right to create their own personal teams for draft/testing purposes.
# 20 Feb 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/20-feb-2025
Archiving
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10603335-20-feb-2025](https://help.textmine.com/en/articles/10603335-20-feb-2025)
It is now possible to archive and delete documents in Legislate.
Each team in Legislate now displays an Active and Archived tab to help you better organise your documents.
Documents can be moved to the Archive section by selecting and clicking on the Archive icon.
![Screenshot]()
You can access Archived documents by clicking on the Archived tab and you can either permanently delete a document or move it back to the Active section.
![Screenshot]()
![Screenshot]()
# 21 May 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/21-may-2025
Copy simplifications and permissions
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11418905-21-may-2025](https://help.textmine.com/en/articles/11418905-21-may-2025)
We have added the following permission:
*
Allow users to create a personal team
We have made copy updates
*
"Create default tag" is now: "Manually insert a tag in the plain text"
*
"Edit default settings" is now: "Edit the default settings of a Vault Business Unit"
We have also rename the Settings section under Vault to "Custom Types" to avoid confusion with the Vault Settings of a Business Unit.
![Screenshot]()
# 22 January 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/22-january-2025
Improvements and Feedback validation
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10446926-22-january-2025](https://help.textmine.com/en/articles/10446926-22-january-2025)
Today we have released the following improvements:
*
Download date and Time stamp
*
Validated tags can be incorporated in model improvements (Opt-in required)
## Download date and Time stamp
When downloading documents from Vault, a time stamp of the download date is now visible at the top of the document.
![Screenshot]()
## Tag validation
Validating tags is important for ensuring that the data extracted by Vault is correct and can be used operationally. However, the validation feedback was not previously used to improve the model's accuracy and behaviour. Users can now opt-in to allow TextMine to improve Vault using their validations.
![Screenshot]()
# 25 March 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/25-march-2026
Labels, Entity Extraction and Document processing as images, Chained tags on tables, entities.
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/14185025-25-march-2026](https://help.textmine.com/en/articles/14185025-25-march-2026)
## Labels
You can now provide short-form labels which will be rendered instead of the prompts in the Vault editor, table views or CSV export
![Screenshot]()
## Process text based documents via image models
Typically documents go through a text extraction and question answering pipeline which performs well for most document types. However, the structure of documents with visual elements or complex structures is often lost at the text extraction step which can sometimes impact performance. It is now possible to set the processing to image based for these document types. This means that the image model will treat each page as an image and answer the questions visually.
![Screenshot]()
## Entity Extraction
It is now possible to define entity extraction tags. These tags will resolve all the entities you've defined in the corresponding document.
![Screenshot]()
You can then review the entities in the entities tab, filter by relation or visualise them in a graph.
![Screenshot]()
![Screenshot]()
![Screenshot]()
You can also create chained tags on top of entities or tables extracted from the document.
![Screenshot]()
# 25 September 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/25-september-2025
Form ADV, Vault integrations, Legislate Folders and Tabular data views
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12414209-25-september-2025](https://help.textmine.com/en/articles/12414209-25-september-2025)
## Form ADV and Vault Integrations
Form ADVs can now be directly downloaded from the SEC via TextMine Agents or Vault's new integration page. When you click on Upload Document, there is a new option to use an integration.
![Screenshot]()
You can then select which integration you would like to use.
![Screenshot]()
Once selected, specify the companies you wish to download Form ADVs.
![Screenshot]()
You will then have the option to validate which companies you wish to download.
![Screenshot]()
Once selected, the download will start and the documents will eventually be processed by Vault
![Screenshot]()
![Screenshot]()
## Legislate Folders and Tabular Data Views
Documents in Legislate can now be organised in folders and you can move documents between folders by selecting them.
![Screenshot]()
We also offer a tabular view of the data within the platform. Simply switch to the grid view.
![Screenshot]()
# 26 Feb 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/26-feb-2025
Editing on the preview and improved queues
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10657329-26-feb-2025](https://help.textmine.com/en/articles/10657329-26-feb-2025)
## Editing on the preview
You can now correct tags by highlighting the answer in the pdf preview. After selecting and confirming the answer, you can make additional edits in the Vault card.
![Screenshot]()
This preview editing functionality won't work as well for scans in which case it might be better to use the lasso functionality in the plain text view of the document.
## Improved queues
We now provide more information on the individual status of a document during the Vault processing. This update is particularly useful when uploading documents in bulk.
![Screenshot]()
The available statuses are:
*
Queued: the document has been uploaded
*
Extracting text: extracting the text from the pdf
*
Predicting: extracting the tags from the pdf
*
Processed: the tags have been extracted and are ready for review
![Screenshot]()
# 26 May 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/26-may-2026
Workbench, Workflows, Playbook, Records
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/15264338-26-may-2026](https://help.textmine.com/en/articles/15264338-26-may-2026)
We’re introducing a new set of AI-powered tools designed to help teams move from documents to action more quickly: Workbench, Workflows, Playbooks, and Records.
## Workbench
Workbench is a new space for getting work done with AI across your documents, workflows, playbooks, and records. You can ask questions, upload files, create reusable work sessions, review AI-generated outputs, and keep related work organised in folders.
![Screenshot]()
It is designed to feel like a working area rather than a search box: start with a task, review the output, and turn useful results into next steps.
![Screenshot]()
Workbench can orchestrate the parts of the platform you have access to. For example, ask it to create a workflow or run an agent on a document.
## Workflows
Workflows help teams automate repeatable document processes. A workflow can be run manually, triggered when documents are added or processed, scheduled to run at a set time, or started from an incoming email.
![Screenshot]()
Teams can use workflows to review documents, apply playbooks, look up or update records, send reports, notify colleagues, and keep track of every run from one place.
![Screenshot]()
You can create a workflow manually, by using the Ask Workflow functionality or via Workbench.
## Playbooks
Playbooks let teams turn review policies, checklists, and business rules into reusable document reviews. Create a playbook once, apply it to documents, and review the results consistently across teams and matters.
![Screenshot]()
![Screenshot]()
Playbook runs can be tracked, filtered, and downloaded, including Word outputs where available, making it easier to share findings and keep an audit trail.
## Records
Records create a structured, searchable view of the important business information found in documents. Teams can define the types of records they want to track, such as companies, agreements, people, vendors, or custom business objects.
Records can be created from document evidence, reviewed as candidates, searched, imported or exported, and queried through Ask Records. All you need to do is to create a schema for your chosen Vault and records will automatically be suggested from the document tags.
![Screenshot]()
![Screenshot]()
Each record is designed to stay linked back to the underlying evidence so users can understand where information came from.
![Screenshot]()
**Why This Matters**
Together, these features help teams spend less time manually reviewing documents and more time acting on reliable information. Workbench helps users ask and act, Workflows automate repeatable steps, Playbooks standardise review, and Records turn document knowledge into durable business data.
# 26 Nov 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/26-nov-2025
Auto-assign documents and export improvements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12943378-26-nov-2025](https://help.textmine.com/en/articles/12943378-26-nov-2025)
## Auto-assign
You can now auto-assign documents to yourself at the time of upload.
![Screenshot]()
This works both for uploads and integrations.
## Export enhancements
The PDF export now displays the name of the user who validated the tags and has more accurate tag page matching.
![Screenshot]()
# 27 August 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/27-august-2025
Aggregation
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12097142-27-august-2025](https://help.textmine.com/en/articles/12097142-27-august-2025)
Vault now supports a new type of extraction called Aggregation. This means that Vault will use multiple, independent sections of document to answer questions. Aggregation is useful for "list all X" queries or open ended questions "E.g. What is the effective date?" when the answer can be inferred from multiple independent sections (e.g. definition, signatures and date of last signature).
Aggregation is available for all tags. To set a tag as an aggregation, simply enable the aggregation tag when you create it.
![Screenshot]()
The answer will then have the aggregation symbol in the Vault tag card:
![Screenshot]()
Aggregation tags can be created for default, custom and vault team tags.
# 28 January 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/28-january-2026
Reports
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13545432-28-january-2026](https://help.textmine.com/en/articles/13545432-28-january-2026)
You can now create PDF or Word reports using data extracted from Vault.
Select the documents which should be included in the report in Vault directly and click on the Generate Report.
![Screenshot]()
You'll then be able to create your report request. You can either use an existing template or create a new request from scratch.
![Screenshot]()
![Screenshot]()
You can also save your report request as a template for future reports. Click on Generate Report and the report will start processing.
You can access reports from the reports icon in Vault.
![Screenshot]()
![Screenshot]()
Click on a report to see the results in more detail.
![Screenshot]()
Findings can be copied or you can download versions of the report in Word or PDF.
![Screenshot]()
![Screenshot]()
# 3 July 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/3-july-2025
CSV export improvements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/11703623-3-july-2025](https://help.textmine.com/en/articles/11703623-3-july-2025)
It now possible to customise which columns to export to CSV. After selecting which documents to export to CSV in Legislate, a pop up will appear asking to confirm which columns to export.
![Screenshot]()
Once exported, the CSV will only contain the columns which have been selected.
# 30 January 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/30-january-2025
Order Vault tags, customisation and reprocessing
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10492265-30-january-2025](https://help.textmine.com/en/articles/10492265-30-january-2025)
TextMine has released the following new functionality:
*
Order Vault tags
*
Customise the extraction pipeline
*
Reprocess documents
## Vault tag ordering
When reviewing tags, it is now possible to sort them by date, name and confidence score:
![Screenshot]()
## Extraction customisation
When using Vault, you can now select which parts of the pipeline will be used as part of the extraction. For example, you can enable or disable:
*
TextMine default fields
*
Signature detection
*
Vault tags
![Screenshot]()
![Screenshot]()
Please note that the signature detection can slow down the analysis.
## Reprocess documents
Documents can be re-analysed by Vault by selecting the document and clicking on the reprocess icon.
![Screenshot]()
Reprocessing a document will apply the settings of the Vault. For example, if signature detection was disabled in the first analysis, it will be applied in the reprocessing if it is enabled in the team.
Warning: reprocessing a document will override the fields which have previously been generated and reviewed.
# 30 October 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/30-october-2025
Sharepoint
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12701080-30-october-2025](https://help.textmine.com/en/articles/12701080-30-october-2025)
It is now possible to import files natively from Sharepoint. To do this you will need the following information about your Sharepoint from your Sharepoint/Azure administrator:
*
Client ID
*
Tenant ID
*
Client Secret
*
Site URL
You can save this details in the Vault Settings of the Vault you would like to connect to TextMine:
![Screenshot]()
Test and then save the configuration.
![Screenshot]()
This will then link the Sharepoint with this Vault meaning you'll be able to import files from Sharepoint into this Vault.
To do this, click upload document and select use an integration.
![Screenshot]()
The Sharepoint option will appear if the settings have been configured.
![Screenshot]()
Select which documents you would like to import:
![Screenshot]()
Documents will then be imported into the Vault and go through the usual extraction process.
![Screenshot]()
# 5 December 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/5-december-2025
Reasoning, Anchors and other enhancements
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13015537-5-december-2025](https://help.textmine.com/en/articles/13015537-5-december-2025)
## Chained tags now provide reasoning
The reasoning allows you to ask more complex questions over e
## Anchor Tags
If you would like the tag to be connected to a different page, you can move it with the anchor icon.
![Screenshot]()
You can then scroll or navigate to the page where you would like to place the tag.
![Screenshot]()
This will then set the tag to the corresponding page and this change will reflect both in the page navigation and pdf export.
## Additional Improvements
The release also includes a number of small improvements.
**Improved image model**
The image model is now able to analyse larger images with more complex tasks.
**Bulk send documents from Legislate to Vault**
![Screenshot]()
**Tag counters are now displayed in the tags section.**
![Screenshot]()
**Summary (beta)**
Generate a summary of a document. This functionality is a beta version and for informational purposes and can't be used yet in pdf or CSV exports. It will be improved on significantly over the next couple of weeks.
![Screenshot]()
# 5 March 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/5-march-2025
Edit default tags, document type text and table extraction controls, instruction analysis
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/13940819-5-march-2025](https://help.textmine.com/en/articles/13940819-5-march-2025)
## Document type improvements
### Edit Default Tags
You can now edit default tags by clicking on the 3 dots next to the default tag in document type settings. You can edit both the question and nature of the tag.
![Screenshot]()
### Embedded Table Extraction
You can now control whether you would like table extraction to happen for a specific document type. Previously you could only control this at the vault level.
![Screenshot]()
### Text extraction granularity
Previously you could only control the use of fast or slow text extraction at the vault level.
You can now control whether you would like the text extraction to use more computationally intensive (and slower) text extraction techniques at a document type level. This is useful for prioritising the heavy extraction techniques for document types with known quality issues (e.g. scans) without compromising the processing speed for other document types which don't require the heavy extraction.
![Screenshot]()
## Instruction Analysis
When creating a report, you can now enable an instruction analysis to determine whether the existing tags are sufficient to perform the request. If there are missing tags Vault will generate and process the missing tags to provide a more complete report. The new tags will appear in the Vault documents.
![Screenshot]()
## Simplified Agent Configuration
We have updated the agent creation and configuration to have less steps and bringing the configuration to the same place.
![Screenshot]()
![Screenshot]()
![Screenshot]()
# 6 February 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/6-february-2025
Improved uploading and downloading
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10521717-6-february-2025](https://help.textmine.com/en/articles/10521717-6-february-2025)
## Uploading
TexMine has implemented improved bulk importing functionality which allows you to carry on with your tasks in TextMine whilst the uploads are happening.
![Screenshot]()
![Screenshot]()
![Screenshot]()
A preview of the uploads appears in the bottom right and allows you to carry on with any ongoing tasks within the platform.
## Downloading
Data in Legislate can now be downloaded directly via the selector.
![Screenshot]()
Moreover, you now have the option to download the annotated PDFs for the selected files.
![Screenshot]()
# 6 May 2024
Source: https://docs.textmine.com/archive/intercom/product-updates/6-may-2024
Bug fixes and new documents
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9288839-6-may-2024](https://help.textmine.com/en/articles/9288839-6-may-2024)
TextMine has released the following updates:
*
Scribe bug fixes related to the creation of fields and the importing of word templates
*
Scribe Styling improvements and functionality
*
Support in Vault for 10K and 20F documents
# 6 November 2025
Source: https://docs.textmine.com/archive/intercom/product-updates/6-november-2025
Cover pages, Semantic Tags, General Actions
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/12770752-6-november-2025](https://help.textmine.com/en/articles/12770752-6-november-2025)
## Cover pages
TextMine now includes a cover page with a table of contents listing the questions and corresponding pages they feature on. This table of contents is hyperlinked meaning that if you click on a question it will take you to the corresponding page. The DRAFT label is shown because it was downloaded from Vault. It is removed on downloads from Legislate.
![Screenshot]()
## Semantic Tags
When Vault is looking to extract data from a document, the standard approach is to evaluate answers from multiple sections and then pick the best option. We recently introduced an aggregation tag which would produce an answer using all the answers.
We now offer semantic tags which present answers from all the sections. To create a semantic tag, enable the semantic tag option and process the tag.
![Screenshot]()
You will then see the semantic tags appear on the right with a count of how many were found in the document.
![Screenshot]()
Normal Tags now also offer a navigator to review the sections the model considered to help with the review of tags and alternative.
![Screenshot]()
## General Actions
It is now possible to instruct TextMine agents to interpret and act on data which has been extracted by Vault.
![Screenshot]()
You'll then see a summary of the findings in the results page.
![Screenshot]()
The general actions currently look at all the documents in a Vault but will be expanded to folders and documents in the coming weeks.
# 8 May 2026
Source: https://docs.textmine.com/archive/intercom/product-updates/8-may-2026
Ask Vault
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/15031216-8-may-2026](https://help.textmine.com/en/articles/15031216-8-may-2026)
Ask Vault is a new feature which allows you to interrogate, transform and analyse the data extracted by Vault. You can invoke by entering Q on your keyboard or by clicking on the Ask Vault button.
You can then ask questions about the documents, reports and agents in the vault.
![Screenshot]()
Within the chat you can:
*
generate tables which can then be exported to CSV/XLSX
*
generate exportable plots from those tables
*
query failed documents
*
Assign/unassign documents
*
Reprocess or change a document's type
*
Run reports or agents
![Screenshot]()
Each answer is presented with sources documents in the Vault to facilitate cross-referencing and steps requiring an action (e.g. assignment) will ask you to confirm before performing the action.
![Screenshot]()
![Screenshot]()
You can pop out the chat into a new window for better visibility.
![Screenshot]()
Ask Vault chats are unique to users and limited to the scope of the vault they are in. This first version does not have any chat history. This means that chats are lost when you refresh the page, or switch screen or team.
# December 2, 2024
Source: https://docs.textmine.com/archive/intercom/product-updates/december-2-2024
Improved pagination and submit documents from Legislate to Vault
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/10221032-december-2-2024](https://help.textmine.com/en/articles/10221032-december-2-2024)
In the original pagination release, we were not able to provide pagination for certain documents such as scans. In this release we now offer pagination for any document which is uploaded to Vault.
We have also introduced the ability for documents in Legislate to be resubmitted to Vault. This can be useful if you would like to extract new data from your documents. You can send a document back to Vault by clicking on the 3 dots in Legislate and selecting "Move document to Vault".
![Screenshot]()
# A quick tour of Scribe (Video)
Source: https://docs.textmine.com/archive/intercom/scribe/a-quick-tour-of-scribe-video
Watch this video to learn more about creating templates with Scribe
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9220860-a-quick-tour-of-scribe-video](https://help.textmine.com/en/articles/9220860-a-quick-tour-of-scribe-video)
In this short video, learn how to create templates with Scribe.
For more in-depth tutorials, read our guides on:
*
[Importing Word templates](https://help.textmine.com/en/articles/9210137-how-to-import-word-templates-into-scribe)
*
[Creating fields](https://help.textmine.com/en/articles/9220737-template-fields)
*
[Adding conditional logic](https://help.textmine.com/en/articles/9220328-how-to-add-conditional-logic-into-your-scribe-templates)
# Editing modes in Scribe
Source: https://docs.textmine.com/archive/intercom/scribe/editing-modes-in-scribe
The different editing modes in Scribe
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9215771-editing-modes-in-scribe](https://help.textmine.com/en/articles/9215771-editing-modes-in-scribe)
Scribe is meant to be a collaborative editor which means that you can leave:
*
Comments
*
Requests
*
Flags
This article provides a step by step guide on how to create them and collaborate!
Scribe templates can't be published until all the comments, requests and flags have been resolved!
## Edit mode and Request mode
Editing mode allows you to make live edits to your Scribe templates whereas request mode allows you to track edits and make suggestions or requests.
You can switch between the modes by clicking on the mode button between the publish and save buttons in the top left hand side of the ribbon.
![Screenshot]()
Edits in request mode will be tracked on the right hand side and you can filter between open and closed requests.
![Screenshot]()
A requester can't accept their own requests
In order to invite people to collaborate, click on the pen in the top right.
![Screenshot]()
## Comments
To leave a comment, simply highlight the portion of the text and the tool bar and enter your comment.
![Screenshot]()
Comments are grouped under the comments tab.
![Screenshot]()
You can switch between the comments, requests and flag tabs by clicking on the tab
![Screenshot]()
## Flags
A flag allows you to escalate an aspect of the draft to the rest of your team. Select the text which needs to be flagged and select the flag item in the tool bar. Flags are then grouped under the flags tab.
![Screenshot]()
# How to add conditional logic into your Scribe templates
Source: https://docs.textmine.com/archive/intercom/scribe/how-to-add-conditional-logic-into-your-scribe-templates
How to manage if then type logic in your templates
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9220328-how-to-add-conditional-logic-into-your-scribe-templates](https://help.textmine.com/en/articles/9220328-how-to-add-conditional-logic-into-your-scribe-templates)
Scribe is a smart template editor which allows you to define how a template will be rendered based on conditional logic. This article explains how to implement conditional logic in your templates.
## What is conditional logic?
Document templates often contain conditional logic which will affect how the document is rendered based on the different conditions.
For example, an employment contract will vary based on whether the employee is full-time or part-time or working remotely.
Instead of having multiple templates for each situation, Scribe allows you to define how the template should render in each of these situations so that it's easier to maintain consistency between your templates. This also simplifies the management of your templates as you only need to manage one template.
Finally, configuring a conditional template into a document is much easier for the users as they'll just need to answer a conditional question as opposed to making edits to the templates themselves.
## How to introduce conditional logic in Scribe?
You'll first need conditional fields to introduce conditional into your template. If you don't already have one, click on the field settings icon and create a new field.
![Screenshot]()
When defining the new field, select radio type and specify the different options.
![Screenshot]()
The next step is to specify which text should be displayed when the different options are enabled.
Highlight the text which should display when the first option is selected and select the field and option in the toolbar.
![Screenshot]()
You can then repeat for the second option:
![Screenshot]()
Whilst the text might appear to repeat in the Scribe template, only the selected option will be rendered in Legislate.
# How to import Word templates into Scribe
Source: https://docs.textmine.com/archive/intercom/scribe/how-to-import-word-templates-into-scribe
Import your document templates
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9210137-how-to-import-word-templates-into-scribe](https://help.textmine.com/en/articles/9210137-how-to-import-word-templates-into-scribe)
Scribe allows you to import Word templates so that you can get started a lot faster. To get started click on create template and specify the new template's folder.
![Screenshot]()
A folder allows you to group templates by type or counter-party. You can select an existing folder or create a new one.
Once you've specified the folder, you can then specify the type of the new template (e.g. NDA). You can select an existing type or create a new one. If you create a new one you'll need to confirm additional details about the template.
![Screenshot]()
After this step you can then choose to import a template by dragging and dropping or by connecting to your favourite file storage system.
![Screenshot]()
Once the file has been imported, it will be converted into a Scribe template!
![Screenshot]()
# Template fields
Source: https://docs.textmine.com/archive/intercom/scribe/template-fields
The different types of fields in Scribe
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9220737-template-fields](https://help.textmine.com/en/articles/9220737-template-fields)
Scribe fields allow you to define variable within your templates. Fields can be used to insert values in the text or define [conditional logic](https://help.textmine.com/en/articles/9220328-how-to-add-conditional-logic-into-your-scribe-templates) which will define which text should render based on the selected values.
This article provides an overview of the different types of fields and how to use them.
## How to create a new field
There are 2 ways to create a field. You can either create a field from the text highlight toolbar or by going to the general field settings.
![Screenshot]()
![Screenshot]()
Once you've clicked on the field settings or add new field button, a pop up will appear with a list of fields you've already created from this or previous templates.
![Screenshot]()
To create a new field, click on the **Create new field** button. The Create new field button will then open a pop where you can define the field's parameters.
![Screenshot]()
## The different types of fields and settings
Scribe offers a wide range of settings so that you can truly customise your templates.
### Name
This is a value you can use internally to reference this template field.
For example, **Start Date**.
### Label
This is the value which will be presented to users when they complete the forms to render documents from the Scribe template.
For example, **What is the start date of the agreement?**.
### Placeholder
This is a value which can be used to explain the spirit of the question to the user so that they know how the answer should be formatted or which type of answer the field is expecting.
E.g. **Monday 19 April 2024** or **19/04/2024**
### Group
Field groups allow you to organise your form which will be presented to users when they wish to create instances of the Scribe template.
For example, you might have an **Agreement Dates** group to group all the fields which relate to dates for example the commencement, renewal and expiry dates.
### Type
This is where you select the field's type. This can be any of the following:
![Screenshot]()
### Validation
This allows you to validate the format of the answer. If there is no validation, select **None**.
![Screenshot]()
### Saving and editing the field
Once you have finished creating your field, click on save. You'll then see the new field in the summary. As long as the template has not been published, you can edit and delete fields.
![Screenshot]()
## How to insert a field
Now that you have created your field, select the portion of the text where the value should be inserted. The text will then be highlighted and the field will be generated on the right.
![Screenshot]()
![Screenshot]()
Read our [guide on creating and inserting conditional fields](https://help.textmine.com/en/articles/9220328-how-to-add-conditional-logic-into-your-scribe-templates) into your Scribe templates
##
# How do I log out of TextMine?
Source: https://docs.textmine.com/archive/intercom/technical-faqs/how-do-i-log-out-of-textmine
Steps to log out of TextMine
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9234104-how-do-i-log-out-of-textmine](https://help.textmine.com/en/articles/9234104-how-do-i-log-out-of-textmine)
TextMine will automatically log you out of the platform if your session has been inactive for more than 24 hours.
Alternatively, you can log out by clicking on your profile and clicking on log out.
![Screenshot]()
# A quick tour of Vault (Video)
Source: https://docs.textmine.com/archive/intercom/vault/a-quick-tour-of-vault-video
Learn how to get started with Vault in this 1 minute video
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9220877-a-quick-tour-of-vault-video](https://help.textmine.com/en/articles/9220877-a-quick-tour-of-vault-video)
Watch this quick tour of Vault to start extracting data from your documents using AI.
For more in-depth guides, read our step by step tutorials:
*
[How to use Vault](https://help.textmine.com/en/articles/9177886-how-to-use-vault)
*
[Validating predictions](https://help.textmine.com/en/articles/9209636-validating-predictions-in-vault)
*
[Making corrections](https://help.textmine.com/en/articles/9208466-making-corrections-in-vault)
# An introduction to Vault teams
Source: https://docs.textmine.com/archive/intercom/vault/an-introduction-to-vault-teams
How to manage your Vault imports
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9209265-an-introduction-to-vault-teams](https://help.textmine.com/en/articles/9209265-an-introduction-to-vault-teams)
To get started with Vault, you first need to create a Vault team. You can do this by clicking on the + sign next to the Vault icon in the left-hand sidebar. Upon doing this, a pop up will appear displaying 2 options:
![Screenshot]()
Vault teams are connected to Legislate teams which means that you can either connect a Vault to an existing Legislate team or you can create a new Vault which will be connected to a new Legislate team which will be created at the same time.
## Legislate teams
Vault serves as the import module for extracting data from document imports. You can process documents, make [corrections](https://help.textmine.com/en/articles/9208466-making-corrections-in-vault) and amendments if required and validate the predictions. Once you've validated the data points of the individual documents you can submit them to Legislate.
![Screenshot]()
A document submitted in Legislate can the be filtered, analysed and exported as CSVs.
![Screenshot]()
The reason for separating Vault and Legislate teams is so distinguish verified data (Legislate) and data which needs to be reviewed (Vault).
# Creating manual fields in Vault
Source: https://docs.textmine.com/archive/intercom/vault/creating-manual-fields-in-vault
How to create manual tags in Vault
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9227354-creating-manual-fields-in-vault](https://help.textmine.com/en/articles/9227354-creating-manual-fields-in-vault)
Vault will try and identify as many tags as possible based on the document and its detected type. However, you may wish to create your own tags.
## How to manually create a tag
To manually create a tag, switch to the plain text view of a Vault import and hover over the text which corresponds to the tag you'd like to create.
![Screenshot]()
Once you've selected the field, it will appear under the Created fields tab on the right hand side.
![Screenshot]()
# How to change a document type in Vault
Source: https://docs.textmine.com/archive/intercom/vault/how-to-change-a-document-type-in-vault
Correct a document type when it has been misclassified
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9227362-how-to-change-a-document-type-in-vault](https://help.textmine.com/en/articles/9227362-how-to-change-a-document-type-in-vault)
Vault will use AI to automatically classify a document type based on its substance. You can correct the document type if Vault has made a mistake or not detected the document type.
You can either do this edit from the Vault home screen by clicking on the document type. A list of all the available document types will appear.
![Screenshot]()
The document will then be re-submitted to the model for data extraction after you select the new type.
You can also change a document type within the Vault editor view by clicking on the document type in the top right hand corner.
![Screenshot]()
# Making corrections in Vault
Source: https://docs.textmine.com/archive/intercom/vault/making-corrections-in-vault
How to correct tags and corrections
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9208466-making-corrections-in-vault](https://help.textmine.com/en/articles/9208466-making-corrections-in-vault)
Vault is a large language model which is able to classify documents and answer questions about them.
Whilst Vault has been trained to classify documents and answer questions accurately, it may occasionally make incorrect predictions due to ambiguities in the document or a lack of training for the specific examples. This article explains how to correct Vault predictions.
## Unspecified document type
If the Vault has not detected the type of the uploaded document, you will see a blue unspecified tag. If this is the case, simply click on the tag and select the correct type from the pop up list of document types.
![Screenshot]()
![Screenshot]()
## Incorrect document type
If the model has classified a type incorrectly, click on the document type and select the correct type from the pop up list of document types.
![Screenshot]()
A correction of the document type can also be made from the Vault editor view by clicking on the document type tag.
![Screenshot]()
The document will then be resubmitted for analysis with the correct document type.
## Incorrect predictions
If you need to reject a tag, simply click on the **x** button.
![Screenshot]()
If you need to correct a value, click on the pen and type the correct value.
# Validating predictions in Vault
Source: https://docs.textmine.com/archive/intercom/vault/validating-predictions-in-vault
How to validate document data tags
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9209636-validating-predictions-in-vault](https://help.textmine.com/en/articles/9209636-validating-predictions-in-vault)
Verifying predictions is a key part of Vault. You may choose to [correct](https://help.textmine.com/en/articles/9208466-making-corrections-in-vault) a prediction because it incorrect, partially correct or correct but not presented in your preferred format. After reviewing and correcting documents in Vault, you then to validate the fields before submitting them on Legislate.
Clicking on a Vault import will take you to the review screen. Click on any of the fields and you'll be taken to the portion of the text which the model used to create the tag.
![Screenshot]()
Click on the pen to make a correction and click on the tick when you're happy with the validation. You can make further edits if required by clicking on the pen.
For fields which contain the correct value, click on the tick to validate the answer.
![Screenshot]()
You can also validate all the fields in bulk by clicking on the validate all button on the top right.
![Screenshot]()
The submit button will then be enabled once all the fields have been validated.
![Screenshot]()
# Vault Consultancy Agreement Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-consultancy-agreement-tags
Tags Vault will look to extract when presented with a consultancy agreement
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9251271-vault-consultancy-agreement-tags](https://help.textmine.com/en/articles/9251271-vault-consultancy-agreement-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a consultancy agreement:
What is the agreement date?
What is the commencement date?
What is the work location?
What is the description of the services which will be delivered?
Do the services have an end date ("yes" or "no")?
What is the end date?
What is the time period that each side must give before ending the contract?
Will the services include Deliverables ("yes" or "no")?
Please specify the deliverables
What are the charge for the deliverable?
Can the consultant hire a substitute ("yes" or "no")?
Can the consultant hire a third party to do administrative, clerical or secretarial tasks concerned with the services ("yes" or "no")?
Will the Consultant process personal data on behalf of the Client or its associates ("yes" or "no")?
How long must the Consultant keep the Client’s confidential information secret?
How the consultant will be paid ("hourly rate", "daily rate", "weekly rate" or "monthly rate")?
What is the Consultant's rate?
What is the amount of time the Consultant is expected to contribute to deliver the services?
What is the period over which the consultant is expected to work to contribute to deliver the services?
How frequently invoices can be submitted?
Within how many days of receipt of an invoice will the Client pay the Consultant?
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Vault Invoice Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-invoice-tags
Tags Vault will look to extract when presented with an invoice
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9259752-vault-invoice-tags](https://help.textmine.com/en/articles/9259752-vault-invoice-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with an invoice:
Unique number (identifier) of given invoice
Name of the organization that supplies the service (supplier)
Client who pays the invoice
Currency of the invoice
Total cost of the invoice
Start Date of the service
End date of the provided service (if provided)
Payment's frequency
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Vault Master Service Agreement Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-master-service-agreement-tags
The generic tags Vault will look to extract from a Master Service Agreement
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9251206-vault-master-service-agreement-tags](https://help.textmine.com/en/articles/9251206-vault-master-service-agreement-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a supplier agreement:
What is the commencement date?
What is the end date?
Can the agreement be terminated ("yes" or "no")?
How can the agreement be terminated?
What is the value of this contract?
Are there consequences for breaching the contract ("yes" or "no")?
What are the consequences for breaching the contract?
Does the supplier provide artificial intelligence services to the client?
Does the supplier use artificial intelligence to deliver its services to the client?
Is the supplier allowed to use the client's data for the purposes of training its artificial intelligence models?
Where is the client's data stored?
Is the Supplier making any representations ("yes" or "no")?
What are the scope of services?
What is the Business Continuity Plan?
Within what timeframe will the Supplier restore services following any disruptions in accordance with the Business Continuity Plan?
How often is the Business Continuity Plan tested?
What laws must the parties comply to in this Agreement?
What record keeping and reporting obligations does the Supplier have?
Who owns the Intellectual Property Rights for the Created IPR?
Who owns the Intellectual Property Rights for the Background IPR?
What is the Statement of Work (SOW) process?
What is the Service Level?
How can a change to the Scope of Services be requested?
What is the pricing schedule?
Can the Supplier pass on additional expenses?
Is the price inclusive of VAT?
What will the Supplier indemnify the Customer against?
Do indemnifications from TUPE/ARD apply?
What happens in the event the Customers use of Services/Deliverables infringe with IPR's?
Can confidential information be disclosed to the Receiving Parties Representatives?
What is the limit of liability in the event of a breach of confidentiality?
What is the limit of liability with regards to indemnification?
What is the limit of liability with regards to Data Protection?
Is there any liability for consequential loss or damage?
What is the Customer's total aggregate liability to the Supplier?
What is the Supplier's total aggregate liability to the Customer?
What insurance level is the Supplier required to maintain?
What are the parties rights with regards to Force Majeure?
What is the Term of the Agreement?
Can the Agreement be terminated without cause?
Can a SOW be terminated without cause?
Can the Agreement be terminated with cause by the Customer?
Can the Agreement be terminated with cause by the Supplier?
Do cross termination rights apply?
Are any fees refundable on the event of termination?
What happens once a written request for termination has been submitted by the Customer?
Can the Supplier subcontract any of its services?
What are the audit rights?
Who bears the cost of the audit?
How can a notice be given under this Agreement?
What is the governing law?
In the event of a conflict which agreement is followed?
What is the dispute resolution procedure?
To suggest additional tags, use Vault's custom tag functionality or get in touch with our team at [support@textmine.com](mailto:support@textmine.com)
# Vault NDA Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-nda-tags
Tags Vault will look to extract when presented with a NDA
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9251241-vault-nda-tags](https://help.textmine.com/en/articles/9251241-vault-nda-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a non-disclosure agreement:
What is the agreement date?
What is the commencement date?
What is the confidentiality term of the agreement?
Does the NDA have a purpose ("yes" or "no")?
What is the purpose of the confidentiality agreement?
Is the non-disclosure agreement "one-way", "mutual" or "triparte"?
Does the confidentiality agreement include a non-compete ("yes" or "no")?
What is the jurisdiction of the agreement?
What is the governing law of the agreement?
What confidential information is being disclosed?
What is the cost of a breach?
How long does the Recipient have to destroy information?
Do any clauses survive Termination ("yes" or "no")?
Does the Discloser retain the rights to the Confidential Information ("yes" or "no")?
Is there an indemnity clause ("yes" or "no")?
Does the recipient of confidential information have Representatives ("yes" or "no")?
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Vault Order Form Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-order-form-tags
Tags Vault will look to extract when presented with an Order Form
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9259772-vault-order-form-tags](https://help.textmine.com/en/articles/9259772-vault-order-form-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with an Order Form:
Unique number (identifier) of given invoice
Name of the organization that supplies the service (supplier)
Client who pays the invoice
Currency of the invoice
Total cost of the invoice
Start Date of the service
End date of the provided service (if provided)
Payment's frequency
Purchased service
List price of the purchased service
Discount applied to the price
Discount
Is there an auto-renewal?
How much notice does the client need to give to opt out of the auto-renewal?
What are the payment terms?
Does a service level agreement apply to the order?
What are the service level agreement terms?
Does the order form reference a purchase order number?
Purchase order number
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Vault Purchase Order Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-purchase-order-tags
Tags Vault will look to extract when presented with a PO
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9259764-vault-purchase-order-tags](https://help.textmine.com/en/articles/9259764-vault-purchase-order-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a purchase order:
Unique number (identifier) of given order
Name of the organization that supplies the service (supplier)
Client who pays the order
Currency of the order
Total cost of the order
Start Date of the service
End date of the provided service (if provided)
Payment's frequency
Status of the purchase order ("approved", "approved not sent", "close", "review", "sent")
Delivery status of the purchase order ("past due", "on time")
Person who requested the purchase order
Vendor contact email
Received quantity
Unreceived quantity
Has the vendor viewed the purchase order ("yes" or "no")?
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Vault Service Agreement Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-service-agreement-tags
Tags Vault will look to extract when presented with a service agreement
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9251255-vault-service-agreement-tags](https://help.textmine.com/en/articles/9251255-vault-service-agreement-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a service agreement:
What is the commencement date?
What is the end date?
Can the agreement be terminated ("yes" or "no")?
How can the agreement be terminated?
What is the value of this contract?
Are there consequences for breaching the contract ("yes" or "no")?
What are the consequences for breaching the contract?
Does the supplier provide artificial intelligence services to the client?
Does the supplier use artificial intelligence to deliver its services to the client?
Is the supplier allowed to use the client's data for the purposes of training its artificial intelligence models?
Where is the client's data stored?
Is the Supplier making any representations?
What are the scope of services?
What is the Business Continuity Plan?
Within what timeframe will the Supplier restore services following any disruptions in accordance with the Business Continuity Plan?
How often is the Business Continuity Plan tested?
What laws must the parties comply to in this Agreement?
What record keeping and reporting obligations does the Supplier have?
Who owns the Intellectual Property Rights for the Created IPR?
Who owns the Intellectual Property Rights for the Background IPR?
What is the Statement of Work (SOW) process?
What is the Service Level?
How can a change to the Scope of Services be requested?
What is the pricing schedule?
Can the Supplier pass on additional expenses?
Is the price inclusive of VAT?
What will the Supplier indemnify the Customer against?
Do indemnifications from TUPE/ARD apply?
What happens in the event the Customers use of Services/Deliverables infringe with IPR's?
Can confidential information be disclosed to the Receiving Parties Representatives?
What is the limit of liability in the event of a breach of confidentiality?
What is the limit of liability with regards to indemnification?
What is the limit of liability with regards to Data Protection?
Is there any liability for consequential loss or damage?
What is the Customer's total aggregate liability to the Supplier?
What is the Supplier's total aggregate liability to the Customer?
What insurance level is the Supplier required to maintain?
What are the parties rights with regards to Force Majeure?
What is the Term of the Agreement?
Can the Agreement be terminated without cause?
Can a SOW be terminated without cause?
Can the Agreement be terminated with cause by the Customer?
Can the Agreement be terminated with cause by the Supplier?
Do cross termination rights apply?
Are any fees refundable on the event of termination?
What happens once a written request for termination has been submitted by the Customer?
Can the Supplier subcontract any of its services?
What are the audit rights?
Who bears the cost of the audit?
How can a notice be given under this Agreement?
What is the governing law?
In the event of a conflict which agreement is followed?
What is the dispute resolution procedure?
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Vault Supplier Agreement Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-supplier-agreement-tags
Tags Vault will look to extract when presented with a supplier agreement
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9251225-vault-supplier-agreement-tags](https://help.textmine.com/en/articles/9251225-vault-supplier-agreement-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a supplier agreement:
What is the commencement date?
What is the end date?
Does the agreement renew automatically ("yes" or "no")?
How much notice must the client give to cancel the renewal?
What is the name of the product which is being sold?
Is the product a software?
What is the description of the product which is being sold?
What is the contract term?
How frequently will invoices be sent to the customer ("annually", "quarterly", "monthly")?
How much time does the customer have to pay the invoices?
What is the governing law of the agreement?
What is the jurisdiction of the agreement?
What is the total liability for the supplier?
When will the supplier indemnify the customer?
Is any of the supplier's intelectual property "granted", "assigned" or "transferred" to Customer?
Does the agreement contain a force majeure clause ("yes" or "no")?
What are the details of the force majeure clause?
Does the supplier have any publicity rights ("yes" or "no")?
What are the publicity rights of the agreement?
Do any restrictions apply to the customer ("yes" or "no")?
What are the restrictions which apply to the customer?
How many user licences has the Customer licensed from the supplier?
Is the Customer's usage limited by the agreement ("yes" or "no")?
Does the supplier provider professional services to the customer ("yes" or "no")?
What are the professional services provided to the customer?
Are there any warranties in the agreement?
What are the details of the warranties
Does the supplier offer a service level agreement ("yes" or "no")?
What are the details of the service level agreement offered by the supplier?
Does the supplier have a written security policy ("yes" or "no")?
What is the written security policy of the supplier?
If applicable, what is the agreement date?
What is the commencement date?
What is the term of the engagement?
Can the agreement be terminated ("yes" or "no")?
How can the agreement be terminated?
What is the termination notice?
What is the value of this contract?
Are there consequences for breaching the contract ("yes" or "no")?
What are the consequences for breaching the contract?
Does the supplier provide artificial intelligence services to the client?
Does the supplier use artificial intelligence to deliver its services to the client?
Is the supplier allowed to use the client's data for the purposes of training its artificial intelligence models?
Where is the client's data stored?
Is the Supplier making any representations?
What are the scope of services?
What is the Business Continuity Plan?
Within what timeframe will the Supplier restore services following any disruptions in accordance with the Business Continuity Plan?
How often is the Business Continuity Plan tested?
What laws must the parties comply to in this Agreement?
What record keeping and reporting obligations does the Supplier have?
Who owns the Intellectual Property Rights for the Created IPR?
Who owns the Intellectual Property Rights for the Background IPR?
What is the Statement of Work (SOW) process?
What is the Service Level?
How can a change to the Scope of Services be requested?
What is the pricing schedule?
Can the Supplier pass on additional expenses?
Is the price inclusive of VAT?
What will the Supplier indemnify the Customer against?
Do indemnifications from TUPE/ARD apply?
What happens in the event the Customers use of Services/Deliverables infringe with IPR's?
Can confidential information be disclosed to the Receiving Parties Representatives?
What is the limit of liability in the event of a breach of confidentiality?
What is the limit of liability with regards to indemnification?
What is the limit of liability with regards to Data Protection?
Is there any liability for consequential loss or damage?
What is the Customer's total aggregate liability to the Supplier?
What is the Supplier's total aggregate liability to the Customer?
What insurance level is the Supplier required to maintain?
What are the parties rights with regards to Force Majeure?
What is the Term of the Agreement?
Can the Agreement be terminated without cause?
Can a SOW be terminated without cause?
Can the Agreement be terminated with cause by the Customer?
Can the Agreement be terminated with cause by the Supplier?
Do cross termination rights apply?
Are any fees refundable on the event of termination?
What happens once a written request for termination has been submitted by the Customer?
Can the Supplier subcontract any of its services?
What are the audit rights?
Who bears the cost of the audit?
How can a notice be given under this Agreement?
What is the governing law?
In the event of a conflict which agreement is followed?
What is the dispute resolution procedure?
# Vault supported documents and tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-supported-documents-and-tags
Documents and tags Vault has been fine-tuned on
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags)
This article details some of the documents for which [Vault](https://help.textmine.com/en/collections/9145550-vault) has currently been trained. Vault will search for generic tags based on the document type. Users can also specify their own tags. Click on the document type to view its generic tags.
## Which documents can Vault understand?
**Company documents**
Accreditation certificate
10-K
20-F
Articles of Association
Confirmation Statement
Advanced Subscription Agreement
Form ADV
Register of members
SAFE Financing document
Share certificate
Subscription Agreement
Term sheet
Written resolution
**Financial Documents**
[Invoice](https://help.textmine.com/en/articles/9259752-vault-invoice-tags)
[Purchase Order](https://help.textmine.com/en/articles/9259764-vault-purchase-order-tags)
[Utility Bill](https://help.textmine.com/en/articles/9259788-vault-utility-bill-tags)
[Order Form](https://help.textmine.com/en/articles/9259772-vault-order-form-tags)
**HR**
Employment
Company Policy
[Consultancy/Contractor Agreement](https://help.textmine.com/en/articles/9251271-vault-consultancy-agreement-tags)
Stock Option Agreement
Staff Handbook
**Corporate Contracts**
Change request
Letter of intent
[Master service Agreement](https://help.textmine.com/en/articles/9251206-vault-master-service-agreement-tags)
[Service Agreement](https://help.textmine.com/en/articles/9251255-vault-service-agreement-tags)
[Supplier Agreement](https://help.textmine.com/en/articles/9251225-vault-supplier-agreement-tags)
Statement of Work
Partnership Agreement
Recruitment terms of business agreement
[Non-Disclosure Agreement](https://help.textmine.com/en/articles/9251241-vault-nda-tags)
## Adding new documents
We are adding new documents to Vault on a bi-weekly basis, please get in touch with us to suggest new document types at [support@textmine.com](mailto:support@textmine.com)
# Vault Utility Bill Tags
Source: https://docs.textmine.com/archive/intercom/vault/vault-utility-bill-tags
This page was imported from Intercom and may describe older navigation or terminology. Source: [https://help.textmine.com/en/articles/9259788-vault-utility-bill-tags](https://help.textmine.com/en/articles/9259788-vault-utility-bill-tags)
[Vault](https://help.textmine.com/en/collections/9145550-vault) will aim to extract company and signatory information as well as the following generic tags when presented with a Utility Bill:
Unique number (identifier) of given bill
Name of the organization that supplies the service (supplier)
Client who pays the bill
Currency of the invoice
New balance of the bill
Due date of the balance
Charging frequency
Last payment date
Balance of the last bill
Value of the credits
Account number
Client address where the service is supplied
Annual electricity cost for the year
Annual gas cost for the year
Annual water cost for the year
Payment due
What is the fuel mix?
What is the voltage?
What is the capacity?
How many kW were consumed during the day tariff?
How many kW were consumed during the night tariff?
How many kW were consumed during the demand charge tariff?
What was the cost of the day tariff?
What was the cost of the night tariff?
What was the cost of the demand charge tariff?
See all [documents and tags](https://help.textmine.com/en/articles/9251187-vault-supported-documents-and-tags).
# Automate work in TextMine
Source: https://docs.textmine.com/automation/index
Learn how to automate repeatable document work with TextMine workflows, playbooks, reusable agents, triggers, approvals, and run history.
Use this section when a TextMine task should run more than once, follow reusable review logic, or be delegated to an AI agent. Workflows are best for durable automation, playbooks are best for reusable review rules, and agents are best for reusable AI assistants.
## How do I automate a repeatable process?
Start with [Workflows overview](/workflows/overview) to understand triggers, steps, conditions, approvals, integrations, reports, and run history. Use [Create a workflow](/workflows/create-a-workflow) when you are ready to design and test an automation.
## How do I reuse review logic?
Use [Playbooks overview](/playbooks/overview) to apply checklists, policies, negotiation rules, and document review criteria to Vault documents. Use [Create and run a playbook](/playbooks/create-and-run) for setup and execution.
## How do I design reusable agents?
Use [Agents overview](/agents/overview) for guidance on when agents should run, what inputs they need, what outputs they return, and how to design governed agent behavior.
# Manage documents and data in TextMine
Source: https://docs.textmine.com/documents-and-data/index
Learn how to import, organize, extract, validate, publish, report on, curate, and structure document-backed information in TextMine.
Use this section when your TextMine work starts with documents or needs to become structured data. Vault is the main place to import, classify, extract, search, validate, and report on documents. Knowledge Bases help teams curate shared source libraries and ask grounded questions across them.
## How do I manage documents in TextMine?
Start with [Vault overview](/vault/overview) to understand document storage, folders, document types, extraction, Ask Vault, and reports. Use [Tags and extraction](/vault/tags-and-extraction) when you need default tags, custom tags, chained tags, entity tags, image tags, or table extraction.
## How do I build a shared source library?
Use [Knowledge Bases overview](/knowledge-bases/overview) when a team needs a reusable collection of uploaded files, Vault documents, Legislate documents, or business records. Knowledge Bases support folders, member access, source profiles, reusable tags, feedback fields, grounded Q\&A, generated artifacts, and reports.
## How do I turn document evidence into business data?
Use [Records overview](/records/overview) when answers need to become structured, reviewable business records. Use [Schemas and review candidates](/records/schemas-and-candidates) to define record fields and review proposed updates.
## When should I use related document tools?
Use [Legislate overview](/legislate/overview) for validated team-document views, [Scribe overview](/scribe/overview) for smart templates, and [Reports and exports](/reports-and-exports/overview) for shareable evidence-backed outputs.
# Agents, Public API, and MCP
Source: https://docs.textmine.com/getting-started/for-agents
How agents and developers should choose between TextMine Public API V3, agent self-service signup, and the optional TextMine MCP server.
# Agents, Public API, and MCP
TextMine supports two integration styles for agents and developer platforms:
* **Public API V3** is the primary integration path. Use it for signup, vaults, document types, uploads, tasks, records, workflows, agents, and billing.
* **TextMine MCP server** is optional. Use it only when the client wants MCP tool discovery and tool calls instead of direct REST calls.
| Need | Use |
| ------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Product context | This docs site and [Product map](/getting-started/product-map) |
| Agent self-service signup | [Agent self-service signup](/integrations/third-party-agent-signup) |
| Machine-readable onboarding | `https://public-api.textmine.com/.well-known/textmine-agent.json` and `https://public-api.textmine.com/v3/third-party-agents/signup/discovery` |
| TextMine MCP server | [TextMine MCP server](/integrations/textmine-mcp-server) |
| Hosted TextMine MCP endpoint | `https://mcp.textmine.com/mcp` |
| MCP authentication options | `tm_...` TextMine API key for direct/private clients, or OAuth bearer token for enabled marketplace/shared-host app flows. |
| MCP setup guides | [TextMine MCP quickstart](https://developers.textmine.com/docs/textmine-mcp) |
| Current API reference | [developers.textmine.com/reference](https://developers.textmine.com/reference) |
| TextMine CLI | [TextMine CLI](/integrations/textmine-cli) and [CLI quickstart](/integrations/cli-quickstart) |
| Agent instructions for API execution | [TextMine developer For Agents](https://developers.textmine.com/docs/for-agents) |
| ReadMe MCP endpoint | [developers.textmine.com/mcp](https://developers.textmine.com/mcp) |
| ReadMe agent index | [developers.textmine.com/llms.txt](https://developers.textmine.com/llms.txt) |
| Full developer context pack | [Agent Context Pack](https://developers.textmine.com/page/llms-full) |
| Integration and MCP provider catalog | [Integration catalog](/integrations/catalog) |
| Important | Self-service agents do **not** need MCP. They sign up through Public API V3, receive an encrypted `tm_...` service-account API key, and then call Public API V3 directly. |
## Which Path Should I Use?
| Goal | Use | Why |
| ----------------------------------------------------------- | ------------------------ | ----------------------------------------------------------------------------------------------------------------------------- |
| Sign up a new autonomous agent | Public API V3 | Signup is REST-only and does not require MCP. |
| Create the first vault | Public API V3 | `POST /v3/vaults` with a name creates the first vault for the agent organization. The agent does not need to know a `teamId`. |
| Create document types and extraction questions | Public API V3 | `POST /v3/document-types` configures typed processing for a vault. |
| Upload files and read processing status | Public API V3 | `POST /v3/documents` accepts multipart uploads. |
| Send Workbench/task-agent instructions | Public API V3 | Use `/v3/tasks` and `/v3/tasks/{taskId}/messages`. |
| Track free allowance and checkout state | Public API V3 | Use `/v3/agent-billing/status`. |
| Generate an SDK or import REST endpoints into an API client | Public API OpenAPI spec | The OpenAPI spec is the canonical REST contract. |
| Let an MCP-native client discover TextMine tools | TextMine MCP server | Optional wrapper over TextMine capabilities for MCP clients. |
| Search API docs from an MCP-aware docs client | ReadMe/developer hub MCP | This is the documentation MCP endpoint, not the TextMine product MCP server. |
## Self-Service Agent Signup
Use the Public API base URL for the environment. For the UK multitenant deployment:
```text theme={null}
https://public-api.textmine.com
```
Start with machine-readable discovery:
```bash theme={null}
curl https://public-api.textmine.com/.well-known/textmine-agent.json
```
Then create an RSA key pair in the trusted agent runtime and sign up:
```bash theme={null}
openssl genpkey -algorithm RSA -pkeyopt rsa_keygen_bits:2048 -out textmine-agent-private.pem
openssl rsa -pubout -in textmine-agent-private.pem -out textmine-agent-public.pem
curl -X POST https://public-api.textmine.com/v3/third-party-agents/signup/self-service \
-H 'Content-Type: application/json' \
-d '{
"organizationName": "Acme Agent Organization",
"agentName": "Acme Contract Agent",
"contactEmail": "billing@example.com",
"acceptedTerms": true,
"acceptedPrivacy": true,
"credentialDeliveryPublicKey": "'"$(awk 'NF {sub(/\r/, ""); printf "%s\\n", $0}' textmine-agent-public.pem)"'",
"credentialDeliveryAlgorithm": "RSA-OAEP-256"
}'
```
The response includes:
* `credentialRef`: a non-secret reference such as `textmine:api-key:123`
* `apiKeyPrefix`: a display prefix that includes `tm_`
* `encryptedApiKey`: the encrypted service-account API key
* `encryptedApiKeyAlgorithm`
* `encryptedApiKeyEncoding`
* `billingStatus`
* `freeCreditsPerPeriod`
* `setupChecklist`
Decrypt `encryptedApiKey` in the trusted runtime, store the resulting `tm_...` key in a secret store, and keep the raw key out of model-visible prompts, task messages, logs, and transcripts.
Use the key on future Public API calls:
```text theme={null}
Authorization: Bearer tm_...
```
## OpenAPI Spec
Yes: provide a Public API V3 OpenAPI spec and link it from the developer hub, agent guide, and MCP docs. Agents and developers use the OpenAPI spec for REST endpoint discovery, SDK generation, typed clients, API-client imports, and exact request/response shapes.
Use the published API reference at [developers.textmine.com/reference](https://developers.textmine.com/reference). The checked-in source for the agent self-service surface is `docs/openapi/textmine-public-api-v3-agent-self-service.openapi.json`.
The Public API OpenAPI spec should include the full self-service bootstrap path:
* `GET /.well-known/textmine-agent.json`
* `GET /v3/third-party-agents/signup/discovery`
* `POST /v3/third-party-agents/signup/self-service`
* `GET /v3/agent-billing/status`
* `POST /v3/vaults`
* `POST /v3/document-types`
* `POST /v3/documents`
* `POST /v3/tasks`
* `POST /v3/tasks/{taskId}/messages`
MCP is different. MCP clients discover runtime tools through `initialize`, `tools/list`, and `tools/call`. Connector platforms may need their own OpenAPI wrapper around `/mcp`, but that wrapper is not the canonical Public API spec.
## REST-First Bootstrap Flow
After signup:
1. Check allowance and setup:
```bash theme={null}
curl https://public-api.textmine.com/v3/agent-billing/status \
-H "Authorization: Bearer $TEXTMINE_API_KEY"
```
2. Create the first vault:
```bash theme={null}
curl -X POST https://public-api.textmine.com/v3/vaults \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"name":"Agent Vault"}'
```
3. Create a document type:
```bash theme={null}
curl -X POST https://public-api.textmine.com/v3/document-types \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"name": "Commercial Agreement",
"vaultId": 123,
"defaultTags": [
{
"name": "Summary",
"question": "Summarize this document in one sentence."
}
]
}'
```
4. Upload a file:
```bash theme={null}
curl -X POST "https://public-api.textmine.com/v3/documents?vault_id=123&document_type_id=456" \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-F "file=@agreement.pdf"
```
`vault_id` is optional. If the agent omits it, TextMine reuses the first accessible vault or creates the default `Agent Vault` for that API-key organization before uploading the document.
5. Create and message a task:
```bash theme={null}
curl -X POST https://public-api.textmine.com/v3/tasks \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"title": "Review agreement",
"instruction": "Summarize the uploaded agreement and identify missing commercial terms.",
"autostart": true
}'
```
```bash theme={null}
curl -X POST https://public-api.textmine.com/v3/tasks/789/messages \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"message":"Also flag any unusual renewal language.","autostart":true}'
```
6. Re-check billing:
```bash theme={null}
curl https://public-api.textmine.com/v3/agent-billing/status \
-H "Authorization: Bearer $TEXTMINE_API_KEY"
```
Document credits are recorded when document processing completes. Agent-task credits are recorded when an agent task completes. If billing status returns `checkoutRequired=true`, call `POST /v3/agent-billing/checkout` and send the returned Stripe Checkout URL to the billing owner so they can buy a one-time credit pack.
## Optional TextMine MCP Server
The hosted TextMine MCP server is a Streamable HTTP MCP endpoint backed by Public API V3. Use the central setup URL:
```text theme={null}
https://mcp.textmine.com/mcp
```
Direct MCP clients can send the same `tm_...` key used for Public API V3:
```text theme={null}
Authorization: Bearer tm_...
Accept: application/json,text/event-stream
```
Clients that cannot control `Authorization`, such as some connector platforms, can send the key in the alternate header:
```text theme={null}
X-TextMine-Api-Key: tm_...
Accept: application/json,text/event-stream
```
OAuth is also supported for enabled marketplace/shared-host app flows. Those clients should use their OAuth connection flow and send the resulting access token as the MCP bearer token instead of asking users to paste `tm_...` keys into prompts or app manifests.
```text theme={null}
Authorization: Bearer
Accept: application/json,text/event-stream
```
OAuth client registrations can use TextMine's protected-resource metadata at `https://public-api.textmine.com/.well-known/oauth-protected-resource`.
Streamable HTTP requires a session handshake:
1. Send `initialize` to `https://mcp.textmine.com/mcp`.
2. Keep the returned `mcp-session-id` response header.
3. Send `notifications/initialized` with `mcp-session-id`.
4. Call `tools/list` or `tools/call` with `mcp-session-id`.
A direct `tools/call` without `mcp-session-id` is rejected.
## Common Confusions
* `https://developers.textmine.com/mcp` is the developer hub documentation MCP endpoint. It helps MCP-aware clients query API docs.
* The TextMine product MCP server is the hosted endpoint used to call TextMine tools.
* Workbench MCP providers are third-party provider connections inside Workbench. They are not the TextMine product MCP server.
* Agent self-service signup is REST-first. MCP is never required for signup.
* The billing owner email collected during signup is a billing contact, not automatically a human web login.
## Developer Documentation Checklist
Keep these phrases visible in page titles, descriptions, and first-screen copy:
* TextMine Public API
* Public API V3
* OpenAPI spec
* agent self-service signup
* third-party agent API key
* service-account API key
* TextMine MCP server
* MCP setup
* `/.well-known/textmine-agent.json`
Link the agent docs, API reference, and MCP docs to each other so agents can choose the right path without guessing.
# Product map
Source: https://docs.textmine.com/getting-started/product-map
How the current TextMine product areas fit together.
TextMine is organized around a small set of connected work surfaces.
| Area | Use it for |
| ------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Workbench | Ask questions, upload or import documents, orchestrate multi-step work, save artifacts, and work across TextMine capabilities and user-enabled MCP providers. |
| Vault | Store documents, extract tags, review evidence, ask questions over documents, run reports, and trigger document processing. |
| Legislate | Review the validated team-document archive after documents are published from Vault into Legislate. |
| Workflows | Build reusable automations that run manually, from document events, from email, on a schedule, or from intake forms. |
| Playbooks | Convert policies, checklists, and negotiation guidance into reusable document reviews. |
| Records | Turn document evidence into structured business records, schemas, relationships, graphs, and review candidates. |
| Agents | Configure on-demand AI agents that can be run directly, from Workbench, or from Workflows. |
| Scribe | Build and fill smart document templates with fields and conditional logic. |
| Integrations | Import from connected providers, submit to external systems, call APIs/webhooks, and expose TextMine's own MCP server to agent clients. |
## Naming guidance
Use TextMine for the overall platform. Use Legislate when you mean the publication or team-document surface: the validated data archive where documents can appear after being published from Vault.
Vault and Legislate are separate surfaces. A document can be processed in Vault, published to Legislate, and later compared across both, but a Vault search should not be described as a Legislate search unless the user is explicitly asking about published team documents.
## Agent guidance
Agents should read [For agents](/getting-started/for-agents) before acting. This Mintlify site explains product meaning and user workflows. Use [Workbench capabilities](/workbench/capabilities), [Vault and editor capabilities](/vault/capabilities), [Workflow blocks](/workflows/blocks), and [Integration catalog](/integrations/catalog) for detailed product capability maps. The developer hub at [developers.textmine.com](https://developers.textmine.com) provides Public API V3 schemas, operation IDs, ReadMe MCP, and client setup guides.
When an agent needs to execute API-backed work, it should use the developer hub [For Agents](https://developers.textmine.com/docs/for-agents) page and the [API reference](https://developers.textmine.com/reference).
# Quick start
Source: https://docs.textmine.com/getting-started/quick-start
A practical first path through TextMine.
Use this path when you are setting up a new workspace or explaining TextMine to a new user.
## 1. Choose the right surface
* Use Vault when the task starts with documents.
* Use Legislate when the task is about validated documents already published from Vault.
* Use Workbench when the task is exploratory or spans several tools.
* Use Workflows when the task should be repeatable or triggered automatically.
* Use Playbooks when the task is a reusable document review.
* Use Records when the outcome should become structured business data.
* Use Scribe when the outcome is a generated document or template.
Agents should also read [For agents](/getting-started/for-agents) before choosing a surface.
## 2. Add documents to Vault
Upload documents into the relevant Vault, choose or confirm document types, then review extracted fields and tags once processing finishes.
## 3. Ask or automate
Use Ask Vault or Workbench for one-off analysis. Use Workflows for repeatable actions such as importing files, applying playbooks, updating records, sending emails, requesting approvals, or generating reports.
## 4. Turn useful outputs into durable assets
Save recurring instructions as workflows, playbooks, agents, record schemas, or Scribe templates so the same process can be repeated consistently. Publish validated documents from Vault to Legislate when they should become part of the team-document archive.
## Developer path
For API-backed setup, start with [API and developer docs](/integrations/api-and-developer-docs), then use the [Public API V3 quickstart](https://developers.textmine.com/docs/getting-started), [API reference](https://developers.textmine.com/reference), and [ReadMe MCP endpoint](https://developers.textmine.com/mcp).
# TextMine Help Center
Source: https://docs.textmine.com/index
Find answers for TextMine Workbench, Vault documents, workflows, records, integrations, permissions, product updates, and agent/API setup.
Use the TextMine Help Center to choose the right product area and complete document, automation, integration, and administration tasks. Start in Workbench for guided analysis, Vault for document intelligence, Workflows for repeatable automation, and Integrations for API or MCP setup.
## How do I get started with TextMine?
* [Get oriented with the product map](/getting-started/product-map)
* [Follow the quick start](/getting-started/quick-start)
* [Route agent and API work](/getting-started/for-agents)
* [Contact TextMine support](/support)
## How do I choose the right help section?
* [Use Workbench](/workbench/overview) for AI-assisted analysis, reports, agents, and reusable artifacts.
* [Manage documents and data](/documents-and-data) for Vault, Legislate, Records, Scribe, reports, and exports.
* [Automate work](/automation) for workflows, playbooks, and reusable AI agents.
* [Connect TextMine](/integrations/overview) for MCP, API, Zapier, Copilot Studio, and provider imports.
* [Manage access and updates](/admin-and-updates) for permissions, release notes, and archived Intercom help articles.
## Where are the developer and agent docs?
Use [developers.textmine.com](https://developers.textmine.com) for Public API V3, exact schemas, operation IDs, ReadMe MCP, and client setup guides.
Key developer links:
* [Developer home](https://developers.textmine.com/page/developer-home)
* [Public API V3 quickstart](https://developers.textmine.com/docs/getting-started)
* [API reference](https://developers.textmine.com/reference)
* [Developer For Agents](https://developers.textmine.com/docs/for-agents)
* [ReadMe MCP endpoint](https://developers.textmine.com/mcp)
* [ReadMe llms.txt](https://developers.textmine.com/llms.txt)
* [Agent context pack](https://developers.textmine.com/page/llms-full)
## Where are older Intercom articles?
The original Intercom articles are preserved in the [Intercom archive](/archive/intercom). Current and updated articles should follow the help-center sections in the sidebar.
# API and developer docs
Source: https://docs.textmine.com/integrations/api-and-developer-docs
Where developers and agents should go for API reference, MCP setup, and exact operation IDs.
Use this page to route developers and agents from product documentation into the API execution layer.
TextMine product documentation lives at this site. API reference, ReadMe MCP, exact schemas, operation IDs, and client setup guides live at [developers.textmine.com](https://developers.textmine.com).
## Developer hub links
| Need | Link |
| ------------------------ | ---------------------------------------------------------------------------------------------------- |
| Developer home | [developers.textmine.com/page/developer-home](https://developers.textmine.com/page/developer-home) |
| Public API V3 quickstart | [developers.textmine.com/docs/getting-started](https://developers.textmine.com/docs/getting-started) |
| API reference | [developers.textmine.com/reference](https://developers.textmine.com/reference) |
| Agent instructions | [developers.textmine.com/docs/for-agents](https://developers.textmine.com/docs/for-agents) |
| MCP quickstart | [developers.textmine.com/docs/textmine-mcp](https://developers.textmine.com/docs/textmine-mcp) |
| ReadMe MCP endpoint | [developers.textmine.com/mcp](https://developers.textmine.com/mcp) |
| ReadMe llms.txt | [developers.textmine.com/llms.txt](https://developers.textmine.com/llms.txt) |
| Agent context pack | [developers.textmine.com/page/llms-full](https://developers.textmine.com/page/llms-full) |
## How to access the API
Use Public API V3 for new integrations and agent work.
1. Confirm the user has access to the relevant TextMine organization, team, Vault, provider, or workflow.
2. Create or request an API credential through the TextMine admin/developer access flow for that organization.
3. Store the credential as an environment variable such as `TEXTMINE_API_KEY` or in the secret store for the agent client.
4. Send the credential as a bearer token on API requests.
5. Use the operation IDs, request schemas, and response schemas from the [API reference](https://developers.textmine.com/reference).
6. Poll asynchronous operations after uploads, provider imports, report generation, workflow runs, document reprocessing, and exports.
Use this shape for scripts and examples:
```bash theme={null}
curl "/v3/..." \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-H "Content-Type: application/json"
```
Use the base URL and environment shown in the developer hub or provided by the TextMine team. Do not paste bearer tokens into prompts, tickets, screenshots, source code, or shared documentation.
For command-line workflows, start with [TextMine CLI](/integrations/textmine-cli) and [CLI quickstart](/integrations/cli-quickstart). The CLI uses the same Public API and MCP credentials; it is not a separate permission layer.
## Client setup guides
The developer hub has setup pages for common agent clients:
* [Codex MCP Setup](https://developers.textmine.com/docs/codex-mcp)
* [Claude MCP Setup](https://developers.textmine.com/docs/claude-mcp)
* [Cursor and Windsurf MCP Setup](https://developers.textmine.com/docs/cursor-windsurf-mcp)
* [ChatGPT and OpenAI MCP Setup](https://developers.textmine.com/docs/chatgpt-mcp)
* [TextMine CLI](/integrations/textmine-cli)
* [CLI quickstart](/integrations/cli-quickstart)
* [Copilot Studio Setup](/integrations/copilot-studio)
## MCP access
Use MCP when an agent client should discover and call TextMine tools from documentation or from an MCP-compatible tool surface.
| MCP surface | Use it for |
| ----------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
| TextMine MCP server | TextMine-hosted tools backed by Public API V3. Use `https://mcp.textmine.com/mcp` for product actions where TextMine exposes an MCP tool directly. |
| ReadMe MCP endpoint | Documentation-backed API discovery and execution from [developers.textmine.com/mcp](https://developers.textmine.com/mcp). |
| Workbench MCP providers | In-app provider scan/import flows that let Workbench browse external repositories and import files into a Workbench session. |
For setup, start with [MCP quickstart](https://developers.textmine.com/docs/textmine-mcp), then use the client-specific guide for Codex, Claude, Cursor/Windsurf, ChatGPT/OpenAI, or the [TextMine CLI](/integrations/textmine-cli).
MCP clients can authenticate with a `tm_...` TextMine API key for direct/private setup, or with an OAuth bearer token for enabled marketplace/shared-host app flows.
## How to choose the right layer
Use Mintlify when the question is:
* What is Vault, Legislate, Workbench, Workflows, Playbooks, Records, Agents, Scribe, or Integrations?
* What should users do in the app?
* How do product areas relate to each other?
* Which screenshots, release notes, and user workflows explain the current feature?
Use ReadMe when the question is:
* Which Public API V3 endpoint should I call?
* What is the exact request or response schema?
* Which operation ID should an agent use?
* How do I connect an MCP-compatible client?
* How do I safely execute provider import, report generation, workflow creation, metadata validation, records, or exports?
## API-backed agent pattern
For API-backed work, agents should follow this pattern:
1. Read [For agents](/getting-started/for-agents) for product routing.
2. Read the developer hub [For Agents](https://developers.textmine.com/docs/for-agents) page for API execution rules.
3. Use Public API V3 operation IDs from the API reference.
4. Discover IDs before mutating data.
5. Ask for confirmation before destructive or broad actions.
6. Keep secrets out of logs, prompts, screenshots, commits, and docs.
## Current API focus
New integrations should use Public API V3. V2 remains available as legacy reference, but V3 is the agent-friendly surface for current public API work.
# Integration catalog
Source: https://docs.textmine.com/integrations/catalog
Supported integrations, Zapier automations, TextMine MCP, agent self-service signup, and provider import capabilities.
TextMine integrations fall into several shapes: OAuth-connected providers, Vault-level native integrations, workflow integrations, Zapier automations, the TextMine MCP server, agent self-service signup, and Workbench MCP/provider imports.
Availability depends on organisation configuration, enabled features, user connection status, and provider capability support.
## Integration directions
| Direction | What it means |
| ---------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------- |
| Import into Workbench | Bring an external file into the current Workbench chat as temporary context. |
| Import into Vault | Save a file into the durable Vault pipeline for processing, extraction, review, workflows, and possible publication to Legislate. |
| Submit out of TextMine | Send outputs, documents, reports, or files to an external system. |
| Expose TextMine to agents | Let external agents call TextMine tools through the hosted TextMine MCP server or Public API V3. |
| Discover API docs | Let agents use the developer hub and ReadMe MCP surface for exact schemas and API operation discovery. |
| Automate with Zapier | Let Zapier triggers, searches, and actions call TextMine through Public API V3. |
| Provision third-party agents | Create a multitenant, non-enterprise agent organization, service-account API key, and billing account through agent self-service signup. |
## Agent and automation entry points
| Entry point | Use it for |
| ------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------- |
| [TextMine MCP server](/integrations/textmine-mcp-server) | MCP-compatible agent clients that need TextMine-hosted tools backed by Public API V3. |
| [Slack](/integrations/slack) | Connect a Slack workspace to TextMine so approved users can ask questions, route work, and use MCP-backed actions from Slack. |
| [Agent self-service signup](/integrations/third-party-agent-signup) | External agents that need a TextMine organization, service-account API key, billing account, Public API V3 access, and MCP access. |
| [API and developer docs](/integrations/api-and-developer-docs) | Public API V3 reference, exact operation IDs, ReadMe MCP, and client setup guides. |
| [TextMine CLI](/integrations/textmine-cli) | Command-line entry point for repeatable Public API, MCP, and agent setup workflows. |
| [For agents](/getting-started/for-agents) | Routing rules for agents deciding between Mintlify, Public API V3, TextMine MCP, ReadMe MCP, Workbench, and the web app. |
| [Zapier](/integrations/zapier) | No-code automations that call TextMine document, record, task, report, workflow, and agent actions through Public API V3. |
| `/.well-known/textmine-agent.json` | Machine-readable agent onboarding metadata for signup, billing, Public API V3, and MCP. |
## User-connected providers
These providers are connected by users or organisations and expose explicit capabilities.
| Provider | Connection shape | Current capabilities |
| ------------ | ------------------------ | ---------------------------------------------------------------------------------------------------------------- |
| Google Drive | OAuth | Create documents, list folders, list files in folders, download files, and import files into Workbench or Vault. |
| SharePoint | OAuth/native integration | List folders, list files in folders, download files, import files, and submit workflow outputs where enabled. |
| Notion | OAuth | Create pages in a connected Notion workspace. |
## Vault native integrations
Vault native integrations are configured against a Vault and can participate in upload/import or workflow processes.
| Integration | What it supports |
| ------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| ServiceNow | Configure a Vault-level ServiceNow connection, test access, list configured table records, list record attachments, and import selected attachments into Vault. |
| DocuSign | Configure, test, enable/disable, and import agreement/envelope material where enabled. |
| SharePoint | Configure, test, browse, import, and submit outputs where enabled. |
| OpenCorporates | Use company data as a source for enrichment or import flows. |
| SEC/EDGAR | Search EDGAR filing text, fetch SEC filings, and use SEC company/adviser sources in upload or workflow flows. |
| Box, Ariba, Finastra, Companies House | Native or legacy import/source surfaces visible in document import flows where configured. |
| Vault email | Route inbound email and attachments into Vault or workflow-triggered document processing. |
## ServiceNow details
The ServiceNow integration can be configured per Vault. Supported behavior includes:
* Basic authentication or OAuth client credentials, depending on configuration.
* Instance URL, table name, encoded query, and display-value settings.
* Default table support such as `ast_contract` where configured.
* Optional field mappings for number, name, supplier, and status display.
* Connection testing before saving.
* Record browsing with search, limit, offset, and optional table override.
* Attachment browsing for selected records.
* Importing selected attachments, selected records, or matching records into Vault.
* Import modes: `UPSERT`, `NEW_RECORDS_ONLY`, and `UPDATE_EXISTING_ONLY`.
Workflow block `import_from_integration` currently supports ServiceNow for pulling records and linked attachments into the workflow Vault. It can use a saved table/query configuration or override the table/query for a run.
## Workflow integration blocks
Workflows can use integration blocks without requiring a human to operate the UI each time.
| Block | Integration behavior |
| ------------------------- | -------------------------------------------------------------------------------------------------------------------- |
| `api_request` | Runs a structured REST request for API integrations, callbacks, and webhooks. |
| `webhook` | Calls an outbound webhook URL or HTTP callback. |
| `import_from_integration` | Pulls records and linked attachments from an external system into the workflow Vault. Currently supports ServiceNow. |
| `submit_to_integration` | Pushes workflow outputs to an external system. Current implementation supports SharePoint output. |
| `search_edgar_filings` | Searches SEC EDGAR filing text and saves matching filings for later steps. |
| `fetch_sec_filings` | Imports recent SEC filings for companies in scope. |
| `send_email` | Sends email notifications with optional output/report/document attachments. |
See [Workflow blocks](/workflows/blocks) for the full workflow catalog.
## Workbench MCP/provider scan and import
Workbench can scan connected providers in a controlled sequence:
1. List enabled providers for the current Workbench session.
2. List folders when the provider supports folder browsing.
3. List or search files in a selected folder.
4. Import selected files into the current Workbench chat.
5. Save imported files into Vault only when the user explicitly asks and approves.
The key capabilities are:
| Workbench capability | What it does |
| --------------------------- | --------------------------------------------------------------------- |
| `list_mcp_providers` | Shows connected providers and capabilities. |
| `list_mcp_folders` | Browses provider folders. |
| `list_mcp_files` | Lists or searches provider files. |
| `import_mcp_documents` | Imports selected files as Workbench context. |
| `import_uploaded_documents` | Saves uploaded or imported Workbench files into Vault after approval. |
Workbench imports selected files rather than scanning an entire external repository. Agents should name the provider, folder, file, or query whenever possible.
## Official MCP/provider catalog
The current provider catalog includes official MCP or source integrations that can be connected manually where enabled. Capabilities are declared per provider.
| Provider | Provider ID | Declared capabilities |
| ---------------- | --------------------------- | ----------------------------------------------------------- |
| Box | `box` | Search documents, list folders, list files, download files. |
| Dropbox | `dropbox` | Search documents, list folders, list files, download files. |
| Amazon S3 | `amazon-s3` | Search documents, list folders, list files, download files. |
| DocuSign | `docusign` | Search documents, download files. |
| iManage | `imanage` | Search documents, list folders, list files, download files. |
| NetDocuments | `netdocuments` | Search documents. |
| HighQ | `thomson-reuters-highq` | Search documents. |
| Ironclad | `ironclad` | Search documents. |
| CourtListener | `courtlistener` | Search documents. |
| CoCounsel Legal | `thomson-reuters-cocounsel` | Search documents. |
| GovInfo | `govinfo` | Search documents, download files. |
| Federal Register | `federal-register` | Search documents, download files. |
| Regulations.gov | `regulations-gov` | Search documents. |
| Congress.gov | `congress-gov` | Search documents. |
| Open States | `open-states` | Search documents. |
| Atlassian | `atlassian` | Search documents. |
| Salesforce | `salesforce` | Search documents. |
| AWS | `aws` | Search documents. |
| Azure | `azure` | Search documents. |
| Google Cloud | `google-cloud` | Search documents. |
| FactSet | `factset` | Search documents. |
| LSEG | `lseg` | Search documents. |
| Stripe | `stripe` | Search documents. |
| Xero | `xero` | Search documents. |
| Snowflake | `snowflake` | Search documents. |
| Databricks | `databricks` | Search documents. |
Manual provider connections store an endpoint URL and optional credential metadata. Secret-like metadata is sanitized and credentials are stored separately. A provider must be connected and Workbench-enabled before Workbench can use it for browsing or import.
## Zapier
Use [Zapier](/integrations/zapier) when a Zap should automate TextMine work through Public API V3. Zapier uses a TextMine service-account API key from agent self-service signup or TextMine provisioning, plus the matching Public API base URL. Open [TextMine on Zapier](https://zapier.com/apps/textmine/integrations) to browse supported pairings and templates.
## TextMine MCP and developer hub
The [TextMine MCP server](/integrations/textmine-mcp-server) exposes TextMine tools to external agents. The [TextMine CLI](/integrations/textmine-cli) gives developers and trusted runtimes a command-line entry point for repeatable Public API, MCP, and setup workflows. The [developer hub](https://developers.textmine.com) provides Public API V3 reference, examples, client setup, and ReadMe MCP discovery.
Use [API and developer docs](/integrations/api-and-developer-docs) when a developer or agent needs exact operation IDs or setup instructions.
# CLI quickstart
Source: https://docs.textmine.com/integrations/cli-quickstart
Configure the TextMine CLI for safe Public API and MCP workflows.
Use this quickstart to prepare a terminal, CI job, or trusted agent runtime for TextMine command-line work.
This page focuses on configuration and first safe checks. Use the install command and command reference supplied with the CLI release or the developer hub for the exact command names in your version.
## 1. Choose the environment
Use the Public API base URL and MCP URL for the TextMine environment you are allowed to access.
For the public UK multitenant deployment:
```bash theme={null}
export TEXTMINE_PUBLIC_API_BASE_URL="https://public-api.textmine.com"
export TEXTMINE_MCP_URL="https://mcp.textmine.com/mcp"
```
For a dedicated deployment, use the values supplied by TextMine. Do not infer the API URL from a web app URL unless the setup guide explicitly says to do that.
## 2. Choose an authentication method
Use one of the supported authentication paths.
For direct/private automation, store a TextMine API key outside model-visible context:
```bash theme={null}
export TEXTMINE_API_KEY="tm_..."
```
For OAuth-enabled marketplace or shared-host flows, use the OAuth flow from that app or setup guide and store the resulting access token in the runtime's approved secret store. OAuth client registrations can use:
```text theme={null}
https://public-api.textmine.com/.well-known/oauth-protected-resource
```
Do not paste API keys, OAuth tokens, refresh tokens, or client secrets into prompts, screenshots, tickets, source code, shell history, or shared documentation.
## 3. Run read-only checks first
Before running commands that create or update TextMine data, confirm the CLI can complete low-risk checks for the configured environment:
| Check | What it proves |
| ---------------------------- | ------------------------------------------------------------------------------------- |
| Auth/session status | The runtime can find its credential and target environment. |
| API health or account lookup | Public API V3 is reachable and the credential is valid. |
| Vault listing | The credential has the expected TextMine organization and Vault access. |
| MCP tools listing | The hosted MCP endpoint is reachable and the Streamable HTTP session handshake works. |
If any check returns an unexpected organization, empty Vault list, or authentication error, stop and fix the environment or credential before running mutating commands.
## 4. Use the right surface
| Goal | Prefer |
| -------------------------------------------------- | -------------------------------------------------------------------------------- |
| Exact REST operation, schema, or automation script | CLI over Public API V3. |
| MCP-native tool discovery or tool call smoke test | CLI over the hosted TextMine MCP endpoint. |
| Human review, document inspection, or approval | TextMine web app or Workbench. |
| New autonomous agent signup | Public API V3 signup flow; then configure the CLI with the resulting credential. |
## 5. Keep CLI output audit-friendly
For support and audit trails, capture non-secret identifiers in CLI output:
* Organization or environment label.
* Vault ID and document ID.
* Workflow ID and run ID.
* Record schema ID and record ID.
* Task ID and task event IDs.
Do not capture raw credentials. If a command prints a secret by mistake, treat the output as sensitive and rotate the credential if needed.
Next, read [TextMine MCP server](/integrations/textmine-mcp-server) for MCP setup details or [Agent self-service signup](/integrations/third-party-agent-signup) for service-account onboarding.
# Set up TextMine in Copilot Studio
Source: https://docs.textmine.com/integrations/copilot-studio
Connect a Microsoft Copilot Studio agent to TextMine from Azure Marketplace with the central TextMine MCP.
Use this guide when a Microsoft Copilot Studio agent should search TextMine Vaults, ask questions over documents, inspect business records, or create and follow task-agent work.
TextMine is available to Copilot Studio through Azure Marketplace as **TextMine MCP**. The marketplace listing connects Copilot Studio to the central hosted TextMine MCP server. The MCP server is backed by Public API V3, so the agent acts with the same TextMine permissions as the OAuth user or API key used for the connection.
Add TextMine from Copilot Studio as a Model Context Protocol tool. The current setup path is Azure Marketplace, Copilot Studio, and TextMine MCP; separate Power Apps setup is not needed.
Customer-managed Copilot Studio setup can use an API-key connection with `X-TextMine-Api-Key`. OAuth-based Microsoft marketplace flows use the same central MCP URL so TextMine can resolve the user, organization, and environment server-side.
## Which setup path should I use?
| Situation | Use this path |
| --------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| The agent's **Tools** page shows **New tool** > **Model Context Protocol** and **TextMine MCP** is listed | Select **TextMine MCP** and create or authorize the connection in Copilot Studio. |
| A TextMine MCP connection already exists in the same Copilot Studio environment | Add the existing TextMine MCP tool to the agent. |
| **TextMine MCP** is not listed | Confirm the selected environment can use Azure Marketplace and MCP tools, then ask your Microsoft administrator or TextMine support to make **TextMine MCP** available. |
| Copilot Studio asks for a server URL | Use the central TextMine MCP endpoint: `https://mcp.textmine.com/mcp`. |
Think of Copilot Studio as the place where the agent gets TextMine as a tool. The TextMine MCP Azure Marketplace listing handles the Copilot Studio connection path.
For general Microsoft setup details, see Microsoft's guides for [connecting an existing MCP server to a Copilot Studio agent](https://learn.microsoft.com/en-us/microsoft-copilot-studio/mcp-add-existing-server-to-agent) and [adding MCP tools to an agent](https://learn.microsoft.com/en-us/microsoft-copilot-studio/mcp-add-components-to-agent).
## Before you start
You need:
* Access to Microsoft Copilot Studio and the environment where the agent lives.
* Permission to add Azure Marketplace and MCP tools to the agent.
* Access to the **TextMine MCP** Azure Marketplace listing in that environment.
* A TextMine OAuth connection or a TextMine API key for the TextMine organization the agent should use, depending on the connection flow. TextMine API keys use the `tm_...` format.
* Confirmation that any required TextMine features, Vaults, integrations, and task-agent email routes have been activated for that organization.
If Copilot Studio asks for a server URL, use the central TextMine MCP endpoint:
```text theme={null}
https://mcp.textmine.com/mcp
```
Do not use `https://developers.textmine.com/mcp`; that is the developer hub MCP endpoint for API reference discovery, not the TextMine product MCP server.
Store TextMine credentials only in the Copilot Studio connection or OAuth vault. Do not paste API keys, OAuth tokens, client secrets, or refresh tokens into agent instructions, prompts, chat transcripts, screenshots, app manifests, or documentation.
## Tutorial: create a new agent and add it to Teams
Use this path when you are starting from a blank Copilot Studio agent and want users to chat with TextMine from Microsoft Teams.
### Create the agent
1. Go to [Copilot Studio](https://copilotstudio.microsoft.com/).
2. Select the environment where the TextMine connection should live.
3. Create a new agent.
4. Give the agent a name and instructions that match the TextMine work it should support.
5. Save the agent.
![Create a new agent in Copilot Studio]()
### Add TextMine MCP as a tool
1. In the saved agent, select **Add a tool**.
2. Select **New tool** if Copilot Studio asks whether to create or reuse a tool.
3. Open the **Model Context Protocol** tab.
4. Select **TextMine MCP** from the list.
5. Select **Create connection** when Copilot Studio asks for a connection.
6. Complete the OAuth authorization flow, or paste the TextMine API key supplied by TextMine if Copilot Studio prompts for an API-key connection. TextMine API keys use the `tm_...` format.
7. Create or authorize the connection, then add the TextMine MCP tool to the agent.
8. Review the discovered tools and choose whether the agent can use all TextMine tools or only selected tools.
![Saved agent with Add a tool selected]()
![Model Context Protocol tab with TextMine MCP selected]()
![Create a new TextMine MCP connection]()
![Enter the TextMine API key for the connection]()
If **TextMine MCP** is not listed, follow [Recommended setup: Azure Marketplace TextMine MCP](#recommended-setup-azure-marketplace-textmine-mcp) to check the Copilot Studio setup path and marketplace availability.
### Test the agent
1. Open **Test your agent**.
2. Ask a low-risk TextMine question, such as listing the Vaults the connection can access.
3. Confirm that the answer only uses TextMine data available to the connected OAuth user or API key.
4. If Copilot Studio prompts you to reconnect, open the connection manager from the prompt, reconnect the TextMine MCP connection, then test again.
![Test the agent and reconnect from connection manager if prompted]()
### Publish and add the Teams channel
1. When the test answers look right, select **Publish**.
2. After publishing completes, open the **Channels** tab.
3. Add a channel.
4. Select **Microsoft Teams**.
5. Follow the Teams channel setup prompts to make the agent available in Teams.
6. If the connection manager prompts you to revalidate, revalidate the TextMine MCP connection before continuing.
7. Open the agent in Teams and ask a TextMine question to confirm the Teams experience works end to end.
![Publish the Copilot Studio agent]()
![Open Channels and add a channel]()
![Select Teams as the channel]()
![Revalidate the TextMine MCP connection if prompted]()
![Interact with TextMine in Teams through the agent]()
## Troubleshooting
| Issue | Check |
| -------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **TextMine MCP** is not listed | Confirm the selected Copilot Studio environment can use Azure Marketplace and MCP tools. Ask your Microsoft administrator or TextMine support to make **TextMine MCP** available. |
| Copilot Studio asks you to reconnect or revalidate | Open the connection manager from the prompt, reconnect or revalidate the TextMine MCP connection, then test the agent again. |
| Authentication fails | Confirm the connection uses `https://mcp.textmine.com/mcp`. For OAuth setup, reauthorize the TextMine connection. For API-key setup, confirm Copilot Studio sends `X-TextMine-Api-Key` and stores the raw `tm_...` key value. |
| Vaults, documents, records, or tasks are missing | Confirm the connected OAuth user or TextMine API key has access to that data and that required TextMine features, integrations, Vaults, or task-agent email routes are activated for the organization. |
# Integrations overview
Source: https://docs.textmine.com/integrations/overview
Learn how TextMine connects to external systems, user-connected providers, MCP servers, Public API docs, Zapier, SharePoint, and Google Drive.
TextMine has several integration surfaces. Availability depends on the organisation configuration, connected user account, and enabled product features.
For the detailed provider list, ServiceNow behavior, workflow integration blocks, and MCP scan/import catalog, see [Integration catalog](/integrations/catalog).
## Two supported directions
TextMine supports both directions:
* Import from connected providers into TextMine, such as browsing Google Drive or SharePoint and bringing selected files into Workbench or Vault.
* Expose TextMine's own tools to external agent clients through the hosted TextMine MCP server and the developer hub.
## User-connected providers
User-connected providers are OAuth-backed integrations available from Settings, Workbench, and the TextMine MCP server.
| Provider | Current capabilities |
| ------------ | ------------------------------------------------------------------------------------- |
| Google Drive | Create documents, list folders, list files in folders, and download files for import. |
| SharePoint | List folders, list files in folders, and download files for import. |
| Notion | Create pages in a connected Notion workspace. |
## Workbench MCP providers
Workbench can use enabled provider connections to browse external storage and import selected files into the current Workbench chat. This is useful for asking questions over a file before deciding whether it should be saved into Vault.
Provider imports can stop at Workbench context or continue into Vault. Workbench context is useful for one-off analysis. Vault import is the durable path when the file should be processed, classified, tagged, reviewed, and later published to Legislate if validated.
## Native and workflow integrations
The backend also contains native integration services for enterprise and document workflows, including Box, DocuSign, SharePoint vault integrations, Ariba, Finastra, OpenCorporates, SEC company filings, ServiceNow, and Vault email processing.
Workflow steps can also call generic REST APIs, send webhooks, import from integrations, submit to integrations, fetch SEC filings, and send email.
## TextMine MCP server
The hosted TextMine MCP server exposes TextMine Public API tools to agent clients. It can work with TextMine data directly and can also use integration tools where the underlying provider connection supports the requested action.
Read [TextMine MCP server](/integrations/textmine-mcp-server) for the product distinction between TextMine MCP, ReadMe MCP, and Workbench MCP providers.
## Developer docs
Use [API and developer docs](/integrations/api-and-developer-docs) when an agent or developer needs exact Public API V3 operation IDs, schemas, ReadMe MCP setup, CLI setup, or client-specific setup guides. Start with [TextMine CLI](/integrations/textmine-cli) for command-line workflows.
# Slack
Source: https://docs.textmine.com/integrations/slack
Connect Slack to TextMine so approved users can ask questions, route work, and use TextMine MCP tools from Slack.
Use this guide when your organization wants to connect Slack to TextMine. The Slack app lets approved Slack users route document questions, task-agent work, and MCP-backed actions to TextMine while TextMine keeps permission checks, audit controls, and organization routing server-side.
TextMine uses one public Slack app and one central MCP endpoint. You do not choose a TextMine environment in Slack. If TextMine needs routing information for your organization, your TextMine administrator enters a TextMine-provided connection value during setup, such as a workspace setup code, organization slug, or approved TextMine URL.
## Before you start
You need:
* Permission to install apps in the Slack workspace.
* A TextMine account with access to the organization, Vaults, workflows, and agents you want to use from Slack.
* A TextMine-provided connection value if your organization requires guided setup.
* Confirmation from TextMine that the Slack app is available for your organization.
If you do not have a setup code or approved TextMine URL, contact your TextMine administrator or TextMine support.
## Install the Slack app
1. Open the TextMine Slack app listing or install link provided by TextMine.
2. Select **Add to Slack**.
3. Review the requested Slack permissions.
4. Approve the installation.
5. When redirected to TextMine, sign in if prompted.
6. Enter the TextMine-provided connection value.
7. Confirm the TextMine organization that should be connected to this Slack workspace.
8. Finish setup.
After setup, TextMine privately resolves the correct routing for your organization. The Slack workspace does not expose a public environment selector.
## Link your Slack user to TextMine
The workspace-level install connects Slack to TextMine. Each person still needs to link their Slack identity to their TextMine user before TextMine can answer with their permissions.
When you first mention TextMine in Slack, TextMine may reply with a link. Open the link, sign in to TextMine, and approve the connection. After linking, TextMine applies your normal TextMine permissions when answering from Slack.
## Use TextMine in Slack
Mention TextMine in a channel or thread where the app is present. Start with low-risk requests:
```text theme={null}
@TextMine search for supplier agreements mentioning renewal.
```
```text theme={null}
@TextMine ask the Commercial Contracts Vault what the termination notice period is in the latest Acme agreement.
```
```text theme={null}
@TextMine create a task to review the uploaded agreement and summarize missing fields.
```
For actions that change data, send messages, create tasks, upload files, trigger workflows, or publish outputs, TextMine agents should explain the planned action and ask for confirmation before acting.
## Use Slack with TextMine MCP
Some Slack and agent surfaces can connect to TextMine through Model Context Protocol (MCP). Use the central TextMine MCP endpoint:
```text theme={null}
https://mcp.textmine.com/mcp
```
For Slack Marketplace or shared-host setup, use the OAuth flow provided by the Slack app or TextMine setup page. TextMine resolves the user, organization, and private routing server-side.
Do not paste TextMine API keys into Slack prompts, public app manifests, screenshots, or shared documentation. Direct/private MCP clients that are not using Slack OAuth may use a `tm_...` TextMine API key, but that key should be stored only in the client connection settings.
## Security model
TextMine keeps Slack, TextMine, and MCP permissions separate:
* Slack workspace installation controls whether the TextMine app can receive events and reply in Slack.
* TextMine account linking controls which TextMine user a Slack user maps to.
* TextMine permissions control which Vaults, documents, workflows, records, and agents can be used.
* MCP authentication controls which TextMine user or service account an external client represents.
A Slack user cannot access TextMine content unless their linked TextMine account has permission to that content.
## Troubleshooting
| Symptom | Check |
| --------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- |
| The Slack app cannot be installed | Confirm you have Slack app-install permission or ask a Slack admin to install it. |
| TextMine asks for a connection value | Enter the setup code, organization slug, or approved TextMine URL provided by TextMine. Do not choose from environment names. |
| TextMine replies with an account-link message | Open the link and sign in to TextMine to connect your Slack identity. |
| TextMine cannot find a Vault or document | Confirm your TextMine account can access that Vault or document in the TextMine app. |
| File download or document upload fails | Confirm the app has permission in the Slack channel and that the file is available to the bot. |
| MCP authentication fails | Use the Slack/TextMine OAuth setup flow for shared-host Slack connections, or store a `tm_...` key only in direct/private MCP client settings. |
For product capability context, see [TextMine MCP server](/integrations/textmine-mcp-server), [Integration catalog](/integrations/catalog), and [Contact support](/support).
# TextMine CLI
Source: https://docs.textmine.com/integrations/textmine-cli
Introduce the TextMine command-line interface for Public API, MCP, and agent setup workflows.
The TextMine CLI is the command-line entry point for developers, operators, and trusted agent runtimes that need repeatable TextMine workflows from a terminal.
Use the CLI when you want to run the same TextMine operation from a local shell, CI job, deployment script, support runbook, or automation runtime. Use the web app when a person needs to inspect documents, review model output, approve changes, or work interactively in Workbench, Vault, Workflows, Records, Playbooks, Scribe, or Legislate.
## What the CLI is for
The CLI should make API-backed work easier without changing TextMine's permission model.
Common uses include:
* Checking which TextMine environment and account a runtime is configured for.
* Running safe read-only checks before an integration goes live.
* Calling Public API V3 operations from scripts.
* Uploading or polling documents in trusted automation.
* Starting and inspecting task-agent work.
* Smoke-testing the hosted TextMine MCP endpoint.
* Capturing reproducible support diagnostics without exposing secrets.
For exact operation IDs and request shapes, use [API and developer docs](/integrations/api-and-developer-docs) and the [Public API reference](https://developers.textmine.com/reference). For agent-facing protocol setup, use [TextMine MCP server](/integrations/textmine-mcp-server).
## How it authenticates
The CLI uses the same credentials as Public API V3 and TextMine MCP. It is not a separate access layer.
| Authentication option | Use it for | Notes |
| ------------------------- | ---------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| `tm_...` TextMine API key | Direct/private CLI use, support-led provisioning, customer-managed automation, and service-account agents. | Store it in a secret manager or local environment variable such as `TEXTMINE_API_KEY`. |
| OAuth bearer token | Marketplace/shared-host app flows and user-authorized environments where TextMine has enabled OAuth. | Use the OAuth flow provided by the app or setup guide. Do not paste tokens into prompts, screenshots, tickets, or source code. |
OAuth client registrations can use TextMine's protected-resource metadata:
```text theme={null}
https://public-api.textmine.com/.well-known/oauth-protected-resource
```
## Environment configuration
Use TextMine-issued environment URLs. For the public UK multitenant deployment, the common values are:
```text theme={null}
TEXTMINE_PUBLIC_API_BASE_URL=https://public-api.textmine.com
TEXTMINE_MCP_URL=https://mcp.textmine.com/mcp
```
Dedicated or customer-specific deployments should use the URLs supplied by TextMine. Do not guess a tenant URL, and do not mix staging credentials with production URLs.
## Safety model
CLI commands can represent real users or service accounts. Treat them like API and MCP tools:
* Read before mutating when current state matters.
* Ask for human confirmation before creating, deleting, publishing, sending, assigning, or triggering workflow actions.
* Keep secrets out of terminal history, logs, prompts, tickets, screenshots, commits, and shared documentation.
* Prefer idempotency keys for retryable create or trigger operations when the API supports them.
* Include Vault, document, workflow, run, record, or task IDs in audit summaries.
Start with [CLI quickstart](/integrations/cli-quickstart) when setting up a new terminal or automation runtime.
# TextMine MCP server
Source: https://docs.textmine.com/integrations/textmine-mcp-server
Set up the central TextMine MCP endpoint for MCP-compatible agent clients.
The TextMine MCP server is the TextMine-hosted Streamable HTTP MCP surface for external agent clients. It lets Codex, Claude, Cursor, Windsurf, ChatGPT, Microsoft Copilot, and automation clients discover and call TextMine tools through an agent-friendly protocol.
Use it when an external agent needs to read or act on TextMine data. Use Workbench when a human is working inside the TextMine app.
## Public endpoint
Use the central TextMine MCP endpoint for public setup guides, MCP registry metadata, and marketplace-style configuration:
```text theme={null}
https://mcp.textmine.com/mcp
```
Use a different TextMine-issued MCP host only when a customer-specific deployment guide explicitly provides one.
Do not use `https://developers.textmine.com/mcp` for TextMine product tool calls. That is the developer hub MCP endpoint for API reference discovery.
## How it fits
TextMine has three agent-facing surfaces that are easy to confuse:
* **TextMine MCP server** exposes TextMine product tools to external MCP clients. It is backed by Public API V3.
* **ReadMe MCP server** is hosted by the developer hub and helps agents discover API documentation and generated API operations from [developers.textmine.com](https://developers.textmine.com).
* **Workbench MCP providers** are in-app integrations that let Workbench browse and import files from connected providers such as storage systems.
For API operation details, examples, and client setup, use [API and developer docs](/integrations/api-and-developer-docs) or the developer hub's [agent guide](https://developers.textmine.com/docs/for-agents). For provider and integration inventory, see [Integration catalog](/integrations/catalog).
## Set up a client
Configure MCP clients with:
```text theme={null}
Server URL: https://mcp.textmine.com/mcp
Transport: Streamable HTTP
```
MCP authentication has two supported patterns:
| Authentication option | Best for | Header |
| --------------------- | ------------------------------------------------------------------------------------------------------- | -------------------------------------------- |
| TextMine API key | Direct/private MCP clients and customer-managed setup. | `Authorization: Bearer tm_` |
| OAuth bearer token | Marketplace apps, shared-host clients, and app-store connection flows where TextMine has enabled OAuth. | `Authorization: Bearer ` |
Direct MCP clients that can send an `Authorization` header can use a TextMine API key as a bearer token:
```text theme={null}
Authorization: Bearer tm_
Accept: application/json,text/event-stream
```
OAuth-enabled app flows can also send an OAuth access token as the bearer token:
```text theme={null}
Authorization: Bearer
Accept: application/json,text/event-stream
```
Use the OAuth flow described by the app or marketplace setup guide. TextMine resolves the user, organization, and environment server-side; users should not paste `tm_...` keys or OAuth tokens into prompts, app package manifests, screenshots, or shared documentation.
Connector platforms that cannot set an `Authorization` header can send the TextMine key in the alternate header:
```text theme={null}
X-TextMine-Api-Key: tm_
Accept: application/json,text/event-stream
```
OAuth client registrations can use TextMine's protected-resource metadata:
```text theme={null}
https://public-api.textmine.com/.well-known/oauth-protected-resource
```
Streamable HTTP clients must create a session before calling tools:
1. Send `initialize` to `https://mcp.textmine.com/mcp`.
2. Keep the returned `mcp-session-id` response header.
3. Send `notifications/initialized` with that `mcp-session-id`.
4. Call `tools/list` or `tools/call` with the same `mcp-session-id`.
A direct `tools/call` without `mcp-session-id` is rejected.
## What it exposes
The MCP server is implemented as a thin wrapper over Public API V3. Each MCP tool should map to an existing V3 endpoint rather than adding new business logic in the MCP service.
Current tool areas include:
* Vaults and Vault documents.
* Document search, create, update, and delete.
* Reports and report templates.
* Workflows and workflow catalogs.
* Agents.
* Alerts.
* Team and Vault users.
* Integrations, provider connection status, provider folder listing, and provider file listing.
When an agent needs exact request or response shapes, route it to the [Public API reference](https://developers.textmine.com/reference). When it needs product semantics, route it back to this Mintlify docs site.
## Import and provider support
TextMine MCP clients can use provider-facing tools where the connected provider supports them. That includes listing integrations, checking provider connection status, listing provider folders or files, and uploading a provider file into Vault through TextMine.
This is separate from Workbench's in-app provider import, but both use the same product concept: external provider files can become TextMine context or TextMine Vault documents.
Use this split when documenting or building agents:
* Import into **Workbench** when the file is temporary context for chat, comparison, summarization, or drafting.
* Import into **Vault** when the file should be processed, extracted, reviewed, governed, used by workflows, or published onward to Legislate.
* Use **Public API V3** when an external system or agent needs a durable integration point.
## Authentication model
The MCP service forwards the authenticated caller to Public API V3. Public API resolves the real user or service-account organization and applies normal TextMine permissions.
The caller can be represented by a `tm_...` TextMine API key or, for enabled marketplace/shared-host app flows, an OAuth bearer token. OAuth callers are resolved server-side before the MCP request reaches Public API V3.
Self-service agents should first get a `tm_...` service-account key through [Agent self-service signup](/integrations/third-party-agent-signup). Enterprise, private, and customer-dedicated deployments should use support-led provisioning unless their setup guide provides an OAuth connection flow.
Do not rely on a shared service user for user-facing MCP clients. Each connection should represent the TextMine user or service account whose Vaults, records, integrations, and tasks the agent may access.
For setup examples, see the developer hub pages for [TextMine MCP](https://developers.textmine.com/docs/textmine-mcp), [Codex](https://developers.textmine.com/docs/codex-mcp), [Claude](https://developers.textmine.com/docs/claude-mcp), [Cursor and Windsurf](https://developers.textmine.com/docs/cursor-windsurf-mcp), and [ChatGPT/OpenAI](https://developers.textmine.com/docs/chatgpt-mcp). For command-line workflows, use [TextMine CLI](/integrations/textmine-cli) and [CLI quickstart](/integrations/cli-quickstart). For Microsoft Copilot Studio, use [Set up TextMine in Copilot Studio](/integrations/copilot-studio).
## Adding a new MCP tool
Add the Public API V3 endpoint first. Then add a small tool class in the MCP service that implements the MCP tools marker interface, injects the Public API client, and exposes one or more tool methods.
Keep provider-specific logic, OAuth flows, database access, validation, and business rules in the API service rather than the MCP service. The MCP layer should stay predictable for agents: tool name, tool description, input schema, API call, response.
## Agent rules
* Prefer named, narrow tools over general-purpose actions.
* Read before mutating when the current state matters.
* Ask for confirmation before creating, deleting, publishing, sending, assigning, or triggering workflow actions.
* Use idempotency keys for retryable create or trigger operations when the API supports them.
* Cite document, Vault, workflow, or run identifiers in agent output so humans can audit the action later.
## When to use it
Use the TextMine MCP server when an external LLM client or automation needs a direct agent-friendly way to read or act on TextMine data. Use the web app when a human is operating interactively inside Workbench, Vault, Workflows, Playbooks, Records, Scribe, or Legislate.
Start with [For agents](/getting-started/for-agents) when choosing between these surfaces.
# Agent self-service signup
Source: https://docs.textmine.com/integrations/third-party-agent-signup
Self-serve onboarding for third-party agents that need a TextMine organization, service-account API key, billing, Public API V3, and MCP access.
**Agent self-service signup** provisions a TextMine organization and service-account API credential for an external agent or automation platform that needs to call Public API V3. The [TextMine MCP server](/integrations/textmine-mcp-server) is optional after the agent has a `tm_...` API key; signup itself is REST-first and does not require MCP.
> **Availability:** self-service signup is only for TextMine's multitenant, non-enterprise deployment. Enterprise, private, and customer-dedicated deployments use support-led provisioning instead of the public self-service flow.
## Discovery and availability
Agents and automation tools should discover signup state before submitting a signup request.
| Endpoint | Purpose |
| --------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| `https://public-api.textmine.com/.well-known/textmine-agent.json` | Public machine-readable discovery document for agent self-service signup, billing, Public API V3, and MCP metadata. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/discovery` | Public API V3 discovery document with the same setup metadata. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/availability` | Returns whether self-service signup is currently available and includes terms, privacy, pricing, and setup checklist metadata. |
For the UK multitenant deployment, use `https://public-api.textmine.com` as the Public API base URL. The commonly used signup URLs are:
| URL | Purpose |
| --------------------------------------------------------------------------- | ------------------------------------------------------------------------- |
| `https://public-api.textmine.com/.well-known/textmine-agent.json` | Public agent discovery document. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/availability` | Signup availability. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/discovery` | Public API V3 signup discovery. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/self-service` | Self-service signup request. |
| `https://mcp.textmine.com/mcp` | Central hosted TextMine MCP endpoint for the same production environment. |
Use the Public API base URL returned by discovery when configuring REST clients. Do not use tenant names or tenant-specific aliases as selectable values for external users; the API key resolves the correct organization.
## Machine-readable onboarding
The discovery response is designed for autonomous onboarding. It tells an agent or trusted runtime:
* Whether agent self-service signup is available.
* Which signup URL and method to use.
* Which signup fields, terms acceptance, and privacy acceptance are required.
* Which credential-delivery algorithm to use for encrypted API-key return.
* Which Public API base URL and MCP endpoint belong to the environment.
* How the `tm_...` API key should be sent after signup.
* Which free-credit, billing-status, checkout, and portal endpoints are available.
* Which setup checklist steps are required before typed document processing.
Current self-service signup does not require pre-authentication or human verification. It can still be unavailable if the deployment switch, tenant switch, or signup protection settings are disabled.
## How signup is enabled
Self-service signup is available only when both are true:
* The deployment-level third-party agent switch is enabled.
* Tenant-level third-party access is enabled.
Backoffice administrators can view and update tenant-level third-party access from Tenant Settings. If signup is unavailable, check the availability message before trying again. Do not bypass the supported signup flow or reuse an API key from another deployment.
## Submit self-service signup
When signup is available, call:
```http theme={null}
POST https://public-api.textmine.com/v3/third-party-agents/signup/self-service
```
Required request fields:
| Field | Description |
| ----------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |
| `organizationName` | Name of the TextMine organization to create for the agent account. |
| `agentName` | Display name used for the service account and service-account API key. |
| `contactEmail` | Billing and operational contact email. |
| `acceptedTerms` | Must be `true`. |
| `acceptedPrivacy` | Must be `true`. |
| `credentialDeliveryPublicKey` | RSA public key controlled by the trusted agent runtime, not the model. Accepts PEM or base64-encoded X.509 SubjectPublicKeyInfo. |
Optional request fields:
| Field | Description |
| ----------------------------- | ---------------------------------------------------------- |
| `credentialDeliveryAlgorithm` | Credential delivery algorithm. Defaults to `RSA-OAEP-256`. |
The response returns:
* `organizationId` and `organizationName`.
* `serviceAccountUserId`.
* `apiKeyId`, `apiKeyPrefix`, and `credentialRef`.
* `encryptedApiKey`, `encryptedApiKeyAlgorithm`, and `encryptedApiKeyEncoding`.
* Billing status, free-credit allowance, and currency.
* `gettingStarted` summary.
* `setupChecklist`.
Decrypt `encryptedApiKey` inside the trusted agent runtime using the private key that matches `credentialDeliveryPublicKey`. Store the decrypted `tm_...` API key in a secret store and inject it into future API or MCP calls. Do not place the raw key in model-visible prompts, logs, plans, or task context. The model should only see `credentialRef`, `apiKeyPrefix`, and non-secret metadata.
Agent task transcripts sanitize secret-like values before persistence. Do not send raw API keys, tokens, passwords, secrets, or credentials in task instructions, messages, context, step input/output, approvals, or events; if they are supplied accidentally, the transcript redactor stores `[REDACTED]` instead.
## What signup creates
When signup succeeds, TextMine creates a third-party agent account with:
* A corporate TextMine organization for the agent.
* The standard feature set needed for agent work: [Vault](/vault/overview), [Scribe](/scribe/overview), [Agents](/agents/overview), [Workflows](/workflows/overview), and [Playbooks](/playbooks/overview).
* An `Agent` business unit and organization role assignment for the service-account actor.
* A service-account user for the third-party agent.
* A `tm_...` service-account API key. For self-service signup, the key is returned as `encryptedApiKey` encrypted to the trusted runtime's public key.
* An agent billing account with the billing contact email, accepted terms/privacy metadata, free-credit allowance, current-period usage, currency, and purchased credit balance.
The billing contact email is the human owner/contact for billing and support. It is not automatically created as a human TextMine login.
## Public API and MCP access
After the trusted runtime decrypts and stores the service-account API key, the third-party agent calls Public API V3 with that credential:
```bash theme={null}
curl "/v3/..." \
-H "Authorization: Bearer $TEXTMINE_API_KEY" \
-H "Content-Type: application/json"
```
The same credential can call the central TextMine MCP server:
```text theme={null}
POST https://mcp.textmine.com/mcp
Authorization: Bearer tm_
Accept: application/json,text/event-stream
```
Use `X-TextMine-Api-Key: tm_` only for connector platforms that own the `Authorization` header. The Public API base URL is the root host where the agent performed discovery/signup, or the deployment-specific host issued by TextMine. Do not append `/v3` when configuring clients such as Zapier; clients append endpoint paths themselves.
Public API resolves the key to the service-account actor and the organization created at signup. The key's organization determines which [Vaults](/vault/overview), [document types](/vault/document-types-and-tags), [workflows](/workflows/overview), [records](/records/overview), [agents](/agents/overview), and other resources the external agent can access.
The discovery metadata also declares an alternate API-key header:
```text theme={null}
X-TextMine-Api-Key: tm_
```
Use the bearer header for Public API and MCP clients unless a platform owns the `Authorization` header.
## First setup steps
The signup response includes a `gettingStarted` object and `setupChecklist`. Use those values to drive the agent's onboarding UI.
Before typed document processing, complete the required steps:
1. Generate a runtime keypair, send `credentialDeliveryPublicKey`, decrypt `encryptedApiKey`, and store the API key outside model context.
2. Accept terms and privacy notices during signup.
3. Create or select a [Vault](/vault/overview), or omit `vault_id` on the first upload so TextMine creates or reuses the default `Agent Vault` for the API-key organization.
4. Configure [document types](/vault/document-types-and-tags), [default tags and extraction questions](/vault/tags-and-extraction) before uploading typed documents.
Optional setup depends on what the agent will do:
* Configure [Playbooks](/playbooks/overview) before running review, redline, or playbook-rule checks.
* Create or select [Workflows](/workflows/overview) before invoking workflow automation.
* Create clear task instructions before using [task agents](/agents/overview).
* Connect MCP clients or the [TextMine CLI](/integrations/textmine-cli) after the API key is stored in the trusted runtime.
## Billing and upgrades
Third-party agent work starts on the configured free-credit allowance for the billing period. Credit costs are returned by the availability and billing-status endpoints, including [Vault document processing](/vault/overview), [workflow execution](/workflows/overview), [Workbench tasks](/workbench/overview), and [agent tasks](/agents/overview).
When the free allowance is exhausted, API-key usage that requires more credits can return `402 Payment Required`. The agent should then:
1. Call `GET /v3/agent-billing/status` to show the current billing status, remaining free credits, purchased credit balance, available credits, checkout requirement, setup checklist, and credit-pack metadata.
2. Call `POST /v3/agent-billing/checkout` to create a Stripe Checkout session.
3. Send the returned checkout URL to the billing owner or show it in the agent's billing UI.
4. After checkout completes and the purchased credits are added to the organization, continue API work.
Use `POST /v3/agent-billing/portal` when an existing billed organization needs a Stripe billing portal session. If checkout has not happened yet, the API can require checkout first.
## Human access
Self-service signup creates the agent's organization and service-account API key. It does not automatically create a human user account for the billing owner.
If a human user needs to access the TextMine web app for the same organization, contact TextMine support with:
* Organization name and organization ID from signup.
* Human user's email address.
* Requested access level or role.
* Whether the user needs API-key management, [Vault setup](/vault/overview), [document type setup](/vault/document-types-and-tags), [workflows](/workflows/overview), or billing support.
After TextMine support links the human user to the organization, the user signs in through the normal TextMine web app. Human users should use their own web session or their own user API key where permitted. Do not share the third-party agent service-account key as a human login credential.
## Service accounts and task-agent email
There are two related service-account concepts:
| Service-account concept | What it is | How it is enabled |
| -------------------------------------- | --------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------- |
| Third-party agent service account | The API actor and `tm_...` key created by third-party agent signup. | Created during signup. |
| Task-agent service account email route | An organization-level inbound email route that lets organization users create and reply to task-agent tasks by email. | Activated by TextMine support for the organization. |
The task-agent email route is separate from the third-party agent API key. It must be activated at the organization level before email can create task-agent work. Third-party agent organizations can have this route too, but it is not automatically enabled just because signup created an API service account.
Contact TextMine support to activate or troubleshoot task-agent email routing for a third-party agent organization.
## Operational notes
* Self-service signup and agent billing are tenant-level capabilities for multitenant, non-enterprise TextMine.
* Enterprise or dedicated deployments should use TextMine support-led provisioning.
* The Public API base URL, central MCP URL, and API key must belong to the same TextMine environment or be routed by TextMine for that environment.
* The billing contact is not the same thing as the service-account API actor.
* Human access, support-managed activation, and task-agent email routing should be requested through TextMine support.
* Agents should use idempotency keys where the API supports them, poll asynchronous work, and show resource IDs in user-visible audit summaries.
# Workbench MCP providers
Source: https://docs.textmine.com/integrations/workbench-mcp-providers
Browse and import connected provider files inside Workbench.
Workbench MCP providers let users bring files from connected external sources into a Workbench conversation.
For the provider list and scan/import catalog, see [Integration catalog](/integrations/catalog).
They are not the same thing as the TextMine MCP server. Workbench MCP providers bring external files into TextMine context. The TextMine MCP server exposes TextMine tools to external agent clients.
## What Workbench can do
Workbench has backend capabilities to:
* List enabled MCP providers.
* List folders from an enabled file-storage provider.
* List files from a selected provider folder.
* Import selected provider files into the current Workbench chat.
Imported files become uploaded-file context for the Workbench task. If the user later asks to save those files into Vault, Workbench uses the normal Vault import pipeline and may require approval.
This means provider import has two natural destinations:
* Import into Workbench for chat context, comparison, summarization, or template filling.
* Import into Vault for durable processing, extraction, review, workflow triggers, and eventual publication to Legislate.
## Supported provider shapes
Provider capabilities are explicit. A provider can support document creation, folder listing, file listing, file download, or a subset of those capabilities.
Google Drive currently supports document creation plus folder/file browsing and download. SharePoint supports folder/file browsing and download. Notion currently supports page creation.
## When Workbench should use MCP providers
Use MCP provider routing when the user explicitly names an external source, connected tool, integration, or MCP provider. Native TextMine requests about Vault, Agents, Workflows, Playbooks, Document Types, Records, or Workbench skills should stay on their native capabilities.
Good Workbench prompts include:
* "Import the latest supplier agreement from Google Drive and compare it with this Vault document."
* "List files in the SharePoint finance folder and bring the lease template into this chat."
* "Use the uploaded file as context, then save the final version into the Finance Vault."
## Agent routing
External agents should treat Workbench MCP providers as a human-in-the-loop app capability. Do not assume that a provider file is a Vault document until it has been explicitly imported into Vault.
For external agent access to TextMine data, use the [TextMine MCP server](/integrations/textmine-mcp-server) or the [developer hub](https://developers.textmine.com). For exact API operations, use the [Public API reference](https://developers.textmine.com/reference).
# Zapier
Source: https://docs.textmine.com/integrations/zapier
Connect TextMine to Zapier for document, record, task, report, workflow, and agent automation through Public API V3.
Use the TextMine Zapier integration when a Zap should read from or write to TextMine through Public API V3.
* Public Zapier page: [TextMine on Zapier](https://zapier.com/apps/textmine/integrations)
* Developer API docs: [developers.textmine.com](https://developers.textmine.com)
* Service-account provisioning: [Third-party agent signup](/integrations/third-party-agent-signup)
Zapier authenticates with a TextMine service-account API key. That key can come from third-party agent signup where self-service signup is enabled, or from TextMine provisioning for an existing organization.
## Choose a setup path
| Situation | What to do |
| -------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| New third-party agent or automation customer | Use [third-party agent signup](/integrations/third-party-agent-signup) when available. Signup provisions an organization, service account, API key, billing account, and getting-started checklist. |
| Existing TextMine organization | Use the service-account API key and Public API base URL issued or shown by TextMine for that organization. |
| Service-account access is not enabled | Use **Contact TextMine** from the Zapier section in TextMine Settings, or ask an administrator to enable third-party access for the tenant. |
Do not create or guess an API key outside the TextMine signup or provisioning flow.
## Find TextMine in Zapier
Open [TextMine on Zapier](https://zapier.com/apps/textmine/integrations), then choose **TextMine** when building or editing a Zap.
TextMine Settings also links to the TextMine Zapier app directory page when service-account access is enabled. Use that route when users need to discover supported triggers, searches, actions, and popular app pairings.
## Discover signup availability
Before asking a new third-party agent to connect Zapier, check whether self-service signup is available:
| Endpoint | Purpose |
| --------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| `https://public-api.textmine.com/.well-known/textmine-agent.json` | Public discovery document for third-party agent signup, billing, API, and MCP metadata. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/discovery` | Public API V3 discovery document with the same setup metadata. |
| `https://public-api.textmine.com/v3/third-party-agents/signup/availability` | Returns whether self-service signup is currently available and includes terms, privacy, pricing, and setup checklist metadata. |
When signup is available, call `POST https://public-api.textmine.com/v3/third-party-agents/signup/self-service` with `credentialDeliveryPublicKey`. The response returns `encryptedApiKey`; decrypt it only inside the trusted runtime or support-managed credential setup, then enter the decrypted service-account API key in Zapier. Do not expose the raw key to model-visible prompts, logs, or shared documentation.
## Find values in TextMine Settings
For existing TextMine organizations, open **Settings** and find the **Zapier** section.
The Zapier section shows:
| Value | What to use it for |
| ---------------------- | --------------------------------------------------------------------- |
| Service account status | Confirms whether Zapier API access is enabled for the organization. |
| Public API V3 URL | Copy this into Zapier as the Public API Base URL. |
| Authentication | Confirms that the connection uses a TextMine service-account API key. |
If service-account access is not enabled for the existing organization, use **Contact TextMine** from the Zapier section.
## Connect TextMine in Zapier
In Zapier, add or reconnect the **TextMine** app.
Enter these fields:
| Zapier field | Value |
| ------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| TextMine API Key | The decrypted service-account API key from third-party agent signup, or the key issued by TextMine provisioning. |
| Public API Base URL | The root Public API URL shown in TextMine Settings or issued during provisioning. Do not add `/v3`. Enter this manually; Zapier should not show preset tenant choices. |
For the UK multitenant deployment, enter `https://public-api.textmine.com` as the Public API Base URL. Do not enter `https://app.textmine.com` or the internal API host.
Zapier tests the connection by calling the Public API health endpoint with the API key. If the key or base URL does not match the service account's environment, authentication fails.
## Available Zapier capabilities
The current Zapier integration is aligned to Public API V3 and includes visible triggers for:
| Trigger | Use it for |
| ------------------------- | ------------------------------------------------------------------------------------ |
| New Document | Start a Zap when a document is uploaded, imported, or processed in a TextMine Vault. |
| New or Updated Agent Task | Start a Zap when a TextMine agent task appears or changes status. |
| New Business Record | Start a Zap when extracted or validated business data appears in a record schema. |
| New Report | Start a Zap when a report appears in a TextMine Vault. |
Searches:
| Search | Use it for |
| ----------------------- | ----------------------------------------------------------------------- |
| Find Document | Retrieve one document by ID. |
| Search Documents | Search documents by query, team, Vault, document type, status, or date. |
| Find Task | Retrieve one agent task by ID. |
| Find Business Record | Retrieve one business record by ID. |
| Search Business Records | Search records in a TextMine record schema. |
| Find Report | Retrieve one report by ID. |
Actions:
| Action | Use it for |
| ------------------------------------ | ------------------------------------------------------------------------------------------------ |
| Import Document From Connected App | Import a file from Google Drive, Box, OneDrive, or SharePoint into TextMine document processing. |
| Update Document | Rename a document, change its document type, or queue reprocessing. |
| Reprocess Document | Queue processing for an existing document. |
| Create Agent Task | Create a TextMine agent task for document review, extraction, or follow-up work. |
| Add Task Message | Add a follow-up instruction or clarification to a task. |
| Start Task | Start or resume a draft or waiting task. |
| Cancel Task | Cancel a non-terminal task. |
| Decide Task Approval | Approve or reject a pending task approval. |
| Create Vault | Create a Vault for an existing team. |
| Create Record Schema | Create a business record schema. |
| Suggest Record Schema | Create a suggested record schema from a prompt or supplied definition. |
| Create Business Record | Create a business record. |
| Create or Replace Record Field Value | Create or replace a field value on a business record. |
| Generate AI Report | Generate a report from a prompt, documents, or a Vault. |
| Create Report Template | Create a custom report template. |
| Create Magic Workflow | Create an enabled Magic Workflow. |
| Create Agent | Create a TextMine Magic Agent. |
| Add Document Assignee | Add an assignee or reviewer to a document. |
Zapier does not create API keys from inside the Zap editor. Use third-party agent signup where available, or contact TextMine when a new existing-organization service-account key is required.
## Tenant and permission behavior
The Public API base URL is an auth/config field so different TextMine deployments can connect to the correct Public API host. Users should use the URL shown in TextMine Settings or issued by TextMine for their service account. Do not prefill this field, and do not present a tenant picker or tenant-name list in Zapier; the service-account API key scopes the connection to its organization.
The service account API key controls the organization, permissions, [Vault](/vault/overview) access, available [records](/records/overview), [workflows](/workflows/overview), and other resources. Entering another customer's Public API host does not grant access to that customer's data.
Keep API keys environment-specific. Do not use a staging key against production, or a production key against another deployment.
## Popular Zap patterns
Common TextMine automations include:
* Send new processed document metadata to Google Sheets.
* Notify Slack when an agent task needs review or approval.
* Update HubSpot, Salesforce, or Pipedrive when TextMine extracts validated business records.
* Generate a TextMine AI report from newly imported documents.
* Import files from a connected storage app into Vault processing.
Use [TextMine on Zapier](https://zapier.com/apps/textmine/integrations) to browse available pairings and templates.
## Troubleshooting
If signup is unavailable:
1. Check `https://public-api.textmine.com/v3/third-party-agents/signup/availability` for the availability message.
2. Confirm third-party access is enabled for the tenant.
3. Confirm self-service signup is enabled for the deployment.
4. Contact TextMine if the tenant or deployment is not configured.
If the Zapier connection fails:
1. Confirm the API key was decrypted from the signup response or issued by TextMine for the correct organization.
2. Confirm the Public API Base URL is the root URL only, without `/v3` or another path.
3. Confirm the key and Public API URL belong to the same TextMine environment.
4. Confirm the key has been decrypted and stored exactly.
5. Contact TextMine if the Settings page says setup is required or if the key has expired.
Never paste API keys into tickets, screenshots, prompts, source code, or shared documentation.
# Knowledge Bases overview
Source: https://docs.textmine.com/knowledge-bases/overview
Collect source material, ask grounded questions, generate artifacts, and manage shared knowledge libraries in TextMine.
Knowledge Bases give teams a shared place to collect source material, ask questions over that material, and turn curated evidence into reusable outputs.
Use a Knowledge Base when the source set is broader than a single Vault folder or when the same collection of materials should support repeated questions, reports, artifacts, or review work.
## When to use a Knowledge Base
| Use Knowledge Bases for | Use another TextMine area when |
| ---------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| Shared libraries of policies, documents, templates, records, or reference material. | A document only needs standard Vault processing and validation. |
| Asking questions across curated sources with source references. | The work is a one-off exploration in Workbench. |
| Creating reusable artifacts, reports, presentations, or review materials from a stable corpus. | The output should run automatically from a trigger; use a Workflow. |
| Turning a drafted review artifact into reusable playbook logic. | The review rules are already known; create or run a Playbook directly. |
## Sources and organization
A Knowledge Base can include uploaded files, Vault documents, Legislate documents, and business records. Teams can organize sources into folders, move documents between folders, and arrange the library with an instruction when a large source set needs structure.
Knowledge Base sources keep their own display names, summaries, document type labels, topics, status, and source metadata. Where source files came from Vault, Legislate, or Records, the Knowledge Base keeps enough source information to let users trace the material back to its origin.
## Source profiles, tags, and feedback
Each source can be profiled before it is used for questions or deliverables. Profiles can include summaries, topics, starter questions, extracted text, tables, entities, and source metadata.
Teams can also add reusable tags to Knowledge Base sources. A tag is a question or extraction target that can be run against a source, stored, and reused for later review. Feedback fields let teams capture structured source-level review data such as status, priority, owner, date, or notes.
Use this when a Knowledge Base needs both searchable context and lightweight review tracking.
## Ask a Knowledge Base
Users can ask a Knowledge Base questions across all ready sources or a selected source subset. Answers are grounded in the hub's source material and can include source references and evidence snippets where available.
Ask is useful for questions such as:
* What does this policy library say about a control or obligation?
* Which documents mention a topic, exception, or risk?
* How do several source documents compare?
* Which source should be used when drafting a report, playbook, or presentation?
## Artifacts and reports
Knowledge Bases can generate artifacts and reports from selected sources. Outputs can include document, presentation, PDF, and report-style deliverables. Reports can include chart data where the source material supports it.
Generated outputs can be refreshed, revised with instructions, versioned, and exported to Word, PowerPoint, or PDF. This makes Knowledge Bases useful for recurring evidence packs, policy summaries, client-ready explainers, internal briefing notes, and draft review assets.
When a Knowledge Base artifact captures review rules or repeatable guidance, it can be promoted into a Playbook so the same logic can be applied consistently to future documents.
## Access and visibility
Knowledge Bases support team and private visibility. Members can be added to a hub, and administrators can manage access for that hub. Use private visibility for work-in-progress or restricted source collections, and team visibility when the hub should become a shared knowledge library.
## Agent guidance
Agents should use a Knowledge Base when the user asks to collect, curate, ask over, or generate from a named source library. Include the Knowledge Base ID, source IDs, report or artifact IDs, and any exported file links in the final answer.
Before making broad changes, confirm the hub, source scope, visibility, and whether generated outputs should stay as drafts, be exported, or be promoted into playbook logic.
# Legislate overview
Source: https://docs.textmine.com/legislate/overview
Use Legislate as the validated team-document archive.
Legislate remains a valid TextMine surface. In the current product model, it is the publication and team-document archive rather than another name for Vault.
## How Legislate relates to Vault
Vault is where documents are uploaded, imported, processed, classified, tagged, reviewed, reprocessed, and analyzed. Once a document or its extracted data is validated, it can be published from Vault to Legislate.
Legislate is where published team documents can be listed, searched, and reviewed as validated archive data. Workbench treats Legislate/team-document inventory as a separate read target from active Vault documents.
## Publication lifecycle
A common lifecycle is:
1. Import or upload the source document into Vault.
2. Assign or confirm the document type.
3. Process the document and extract tags, tables, entities, or image-based evidence.
4. Review and validate extracted values in the Vault editor.
5. Run any required playbooks, approvals, reports, or workflows.
6. Publish the validated document from Vault to Legislate.
7. Use Legislate for archive search, team document review, and comparison with future Vault activity.
## When to use Legislate
Use Legislate when the task is about:
* Published team documents.
* Validated archive data.
* Documents already sent from Vault to Legislate.
* Comparing published archive content with active Vault content.
* Workflows that publish documents or respond to publication events.
## When to use Vault instead
Use Vault when the task is about:
* Document upload or provider import.
* Processing or reprocessing.
* Document types and extraction settings.
* Default tags, custom tags, chained tags, entity tags, image tags, tables, and embedded image detection.
* Ask Vault and reports over active document sets.
* Human validation in the Vault editor.
## Workflows and Legislate
Workflows can publish documents to Legislate after validation or approval. They can also react to publication-related events where configured. Use approval steps before publication when a workflow could move documents into the validated archive.
## Workbench behavior
Workbench routes Legislate inventory questions to Legislate/team-document reading. It should not answer those questions by searching Vault unless the user explicitly asks for Vault, or asks to compare Vault and Legislate.
## Agent guidance
Agents should not assume a Vault document has been published. When the user asks about validated archive data, search or reference Legislate. When the user asks about processing, extraction, tags, or validation, use Vault. When comparing both, name which answer came from Vault and which came from Legislate.
# Modernization report
Source: https://docs.textmine.com/modernization-report
# Documentation Modernization Report
## What changed
* Created a current-product docs structure around Workbench, Vault, Legislate, Workflows, Integrations, Playbooks, Records, Agents, Scribe, and Administration.
* Added generated pages for Legislate, Workbench MCP providers, and the hosted TextMine MCP server based on current frontend/backend code.
* Preserved all imported Intercom screenshots under `/images/intercom`.
* Preserved converted Intercom articles under `/archive/intercom` with a stale-content note.
* Added `llms.txt` so agents have a short map of the current docs.
## Code surfaces used as source
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/routes/secured-routes/secured-routes.module.ts`
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/pages/workflow/workflow-builder/workflow-builder.component.ts`
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/pages/workflow/workflow-step.utils.ts`
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/pages/business-records/business-records.component.ts`
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/pages/playbook/playbook.component.ts`
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/pages/workbench/workbench.component.ts`
* `front-end/textmine-web-workspace/projects/textmine/web/src/app/core/services/integration-providers.service.ts`
* `back-end/services/api/shared/src/main/java/com/legislate/api/workflow/dto/WorkflowActionType.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/workflow/dto/WorkflowTriggerTypeEnum.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/workbench/controller/WorkbenchController.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/workbench/service/task/WorkbenchTaskCapabilityRegistry.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/workbench/service/task/WorkbenchTaskExecutorImpl.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/integrations/providers/IntegrationProvider.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/integrations/providers/FileStorageProvider.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/integrations/providers/DocumentProvider.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/integrations/providers/GoogleDriveDocumentProvider.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/integrations/providers/SharePointIntegrationProvider.java`
* `back-end/services/api/shared/src/main/java/com/legislate/api/integrations/providers/NotionDocumentProvider.java`
* `back-end/services/mcp/src/main/java/com/legislate/mcp/tools/*.java`
* `back-end/services/mcp/src/main/java/com/legislate/mcp/config/PublicApiClientConfig.java`
* `back-end/services/mcp/docs/adding-a-tool.md`
## Source caveats
* `back-end/services/mcp/docs/adding-a-tool.md` contains older authentication wording about a static `MCP_API_KEY`; the generated docs follow the current `PublicApiClientConfig` code, which forwards the caller's bearer Authorization header to Public API.
## Manual review still needed
* Capture fresh screenshots for the current Workbench, Workflow builder, Records, Playbooks, Vault, and Scribe pages.
* Decide whether to publish the Intercom archive publicly or keep it unlisted for migration traceability.
* Add detailed click-by-click tutorials after product owners confirm the current IA and terminology.
* Set up redirects from old Intercom article URLs to the new maintained pages.
# Playbook analytics
Source: https://docs.textmine.com/playbooks/analytics
Use rule-level playbook analytics to monitor review outcomes, identify exceptions, and improve playbook rules.
Playbook analytics turns completed playbook runs into rule-level feedback. Use it to see which rules are consistently satisfied, which rules create exceptions, and where the playbook itself may need clearer wording.
## What analytics shows
The analytics view summarizes each playbook across completed runs. It can show:
* Total and completed runs.
* Compliant, non-compliant, and needs-review rule counts.
* Rules that were not observed in the reviewed documents.
* Potential conflicts between playbook rules.
* Recommendations for improving rule wording or correction guidance.
* Related runs for the selected playbook.
Analytics is most useful after the same playbook has been applied to multiple comparable documents.
## Rule outcome sections
Analytics separates rule outcomes into focused sections:
| Section | What it means |
| ----------------------- | -------------------------------------------------------------- |
| Consistently compliant | Rules that are being satisfied across observed runs. |
| Rules needing attention | Rules with non-compliant, missing, or not-assessable signals. |
| Recommendations | Suggested playbook corrections based on repeated run patterns. |
| Potential conflicts | Rules that may overlap, contradict, or need clearer priority. |
| Not observed | Rules that did not appear in the reviewed document set. |
Use these sections to distinguish real document exceptions from playbook wording issues.
## Recommendations and corrections
When TextMine detects recurring issues, analytics can suggest changes to a playbook rule. A recommendation can explain why a rule needs attention, preview the current wording, and show the proposed correction.
Reviewers can apply, discard, restore, or apply all recommendations. Applied recommendations update the playbook draft so the team can inspect the changes before relying on the revised version.
Do not apply recommendations blindly. Treat them as review assistance for the playbook owner, especially when the playbook governs legal, compliance, credit, procurement, or client-facing work.
## Reviewed document iterations
Where reviewed Word document versions are available, analytics can also show resolution signals. These include resolved non-compliance, average iterations to resolution, and the number of reviewed versions compared.
Use these metrics to understand whether playbook findings are leading to corrected document language over time.
## Export analytics
Playbook analytics can be exported to PDF. Use exports for review packs, governance reporting, quality reviews, or discussions with playbook owners.
When exporting several playbooks, filter the analytics view first so the PDF contains the playbooks relevant to the review.
## Agent guidance
Agents should read analytics before proposing changes to a mature playbook. Summaries should include the playbook ID, run count, high-risk rules, repeated non-compliance, unobserved rules, and any recommendations the user approved.
Ask before applying recommendations or enabling a revised playbook, because those changes affect future document reviews.
# Create and run a playbook
Source: https://docs.textmine.com/playbooks/create-and-run
Create a playbook, apply it to Vault documents, and review run outputs and analytics.
Use this page when creating a reusable document review or applying an existing playbook to Vault documents.
## Create a playbook
Create a playbook from source text, a checklist, a policy, a review guide, a rule table, an uploaded document, or a Knowledge Base artifact. TextMine parses the content and saves it into a reusable playbook.
Good source material includes:
* Preferred clause positions.
* Fallback positions.
* Red flag language.
* Escalation requirements.
* Required fields or clauses.
* Review questions.
* Risk severity guidance.
* Examples of acceptable and unacceptable wording.
## Structure rules clearly
A useful playbook rule should answer:
| Rule part | Example |
| -------------------- | ------------------------------------------------------------------ |
| Topic | Limitation of liability |
| Preferred position | Liability cap equals fees paid in the last 12 months. |
| Fallback position | Higher cap allowed for data breach, confidentiality, or indemnity. |
| Red flag | Unlimited liability for indirect damages. |
| Reviewer action | Escalate to legal if cap is missing or uncapped. |
| Evidence expectation | Cite clause text and page/section where available. |
## Scope and availability
Scope the playbook to the relevant teams, Vaults, or document sets when it should only be available for certain work. Keep draft playbooks disabled until they have been tested on representative documents.
Users with playbook create permissions can edit, enable, disable, or delete playbooks depending on organization policy.
## Run a playbook
Run a playbook against selected Vault documents from:
* The Playbooks page.
* Vault or the Vault editor.
* Workbench.
* A Workflow step.
Use the runs view to inspect status, source, document, output, and comparison details.
## Review outputs
After a run, review:
* Findings and recommendations.
* Source passages and evidence.
* Pass/fail or accepted/deviation status.
* Risk severity.
* Missing clauses or missing evidence.
* Generated Word artifact or review memo where available.
* Comparison details for each reviewed document.
## Review analytics after runs
Use [Playbook analytics](/playbooks/analytics) once a playbook has completed runs. Analytics can show consistently compliant rules, rules needing attention, recommendations, potential conflicts, and rules that were not observed.
Where reviewed Word document versions are available, analytics can also show whether non-compliant findings were resolved and how many reviewed iterations were needed.
## Auto-apply
When configured, playbooks can be applied automatically after Vault document type detection or processing. Workflows can also apply playbooks as part of a larger document process.
Auto-apply is useful when every document of a type should receive the same review, but it should be paired with review queues or approvals for high-risk outputs.
## Agent guidance
Agents should inspect existing playbooks before creating a new one. When proposing a playbook, include rule topics, preferred positions, fallback positions, red flags, output format, and scope. Ask for approval before enabling a playbook, applying it to many documents, using results for publication or external delivery, or applying analytics recommendations.
# Playbooks overview
Source: https://docs.textmine.com/playbooks/overview
Learn how TextMine playbooks apply reusable document review rules, checklists, negotiation guidance, analytics, and structured findings to Vault documents.
Playbooks turn review policies, clause checklists, rule tables, fallback positions, and negotiation guidance into reusable document reviews.
![TextMine Playbooks screen showing reusable review rules for Vault documents]()
Use Playbooks when a team wants consistent review logic across documents. Use Workbench when the review is still exploratory, and use Workflows when the review should run automatically or require approvals, reports, records, or integrations.
## What a playbook contains
A playbook can include:
* Review rules.
* Expected positions.
* Fallback positions.
* Red flags.
* Clause examples.
* Risk severity.
* Recommended actions.
* Reviewer guidance.
* Comparison criteria.
TextMine can parse playbook content from source text, policies, checklists, tables, uploaded documents, or Knowledge Base artifacts and convert it into structured review rules.
## Playbook types
| Type | Use it for |
| ------------------------- | ------------------------------------------------------------------------------------------------------ |
| Contract review playbook | Review clauses in NDAs, MSAs, supplier agreements, credit agreements, leases, or other contracts. |
| Policy playbook | Check documents against a policy or compliance standard. |
| Negotiation playbook | Compare document language against preferred and fallback positions. |
| Risk playbook | Identify red flags, severity, mitigations, and escalation requirements. |
| Master agreement playbook | Compare amendments, order forms, schedules, or statements of work against a master agreement position. |
## Where playbooks run
Playbooks can be run manually, from Workbench, from Vault, after Vault processing, or as a Workflow step. Runs are tracked as playbook applications with status, reviewed documents, results, and comparison details.
Typical run sources:
* A user runs a playbook against selected Vault documents.
* Workbench applies a playbook during supervised analysis.
* A workflow applies a playbook after document import or processing.
* A document type or Vault configuration auto-applies a playbook where enabled.
## Playbook outputs
A playbook run can produce:
* Clause-by-clause findings.
* Risk issue lists.
* Pass/fail or accepted/deviation status.
* Comparison against preferred and fallback positions.
* Review memos or Word artifacts.
* Evidence references to the underlying document.
* Escalation or approval inputs for a workflow.
## Analytics and improvement
Use [Playbook analytics](/playbooks/analytics) after a playbook has completed runs. Analytics shows compliant rules, non-compliant rules, rules needing review, unobserved rules, potential conflicts, and recommendations for improving rule wording or correction guidance.
Analytics helps playbook owners decide whether repeated exceptions are genuine document issues, missing evidence, or playbook rules that should be clarified.
## Example use cases
| Team | Playbook | Output |
| ------------------ | ------------------------ | ---------------------------------------------------------------------- |
| Legal | NDA review | Red flags, fallback positions, and approval recommendations. |
| Procurement | Supplier contract review | Renewal, termination, liability, pricing, and data-protection issues. |
| Financial services | Credit agreement review | Covenant, reporting obligation, security, and jurisdiction exceptions. |
| Risk advisory | Control evidence review | Exceptions, missing evidence, and recommended remediation. |
## Agent guidance
Agents should inspect existing playbooks before creating a new one. If the user wants a repeatable policy review, propose a playbook. If the user wants routing, scheduling, approvals, emails, or integration steps, propose a workflow that applies a playbook.
Before running a playbook across many documents, confirm the scope, Vault, document type, and expected output. Before changing a mature playbook, inspect analytics and ask before applying recommendations.
# Ask, graph, and enrich records
Source: https://docs.textmine.com/records/ask-graph-and-enrichment
Use records as a queryable business knowledge layer.
Records are not just a data store. They are a business knowledge layer that can be queried, visualized, enriched, reviewed, and connected back to source evidence.
## Ask Records
Ask Records answers questions over structured records, source evidence, and related documents.
Use Ask Records for questions such as:
* Which suppliers have high-risk findings and upcoming renewals?
* Which agreements are missing governing law or expiry date?
* Which obligations belong to this counterparty?
* Which records need review?
* What evidence supports this value?
* Which records changed after the latest Vault import?
Good answers should identify the record, field, relationship, and evidence used.
## Graph view
Use the graph view to inspect relationships between records. This is especially useful for connected entities such as companies, agreements, counterparties, obligations, systems, risks, controls, and ownership structures.
Common graph tasks:
* Explore all agreements connected to a supplier.
* Find obligations linked to a contract and responsible party.
* Inspect ownership or affiliate relationships.
* Review which documents support a record.
* Identify missing or unexpected relationships.
## Enrichment
Records can be enriched from configured sources such as web, Wikipedia, SEC, FT Markets, or news sources where enabled. Use enrichment when records need supplemental context beyond the documents already in Vault.
Examples:
| Record type | Enrichment use |
| -------------- | --------------------------------------------------------------------- |
| Company | Jurisdiction, public filings, market context, or profile information. |
| Supplier | External risk signals or ownership context. |
| Fund or issuer | SEC/filing context and public-market references. |
| Risk | News, regulatory, or market context that may affect severity. |
Enrichment should not overwrite validated internal data without review. Treat it as supplemental evidence or candidate input unless the organization has configured a trusted update flow.
## Import and export
Use import/export when records need to move between TextMine and another data system or review workflow.
Typical patterns:
* Import source documents into Vault, then create record candidates.
* Export accepted record data for review or downstream reporting.
* Use workflows to create or update records from validated document evidence.
* Use API or MCP access for controlled system-to-system record operations.
## Agent guidance
Agents should distinguish three kinds of answers:
* **Structured answer**: comes directly from record fields.
* **Evidence-backed answer**: combines records with source documents, tags, or passages.
* **Enriched answer**: includes external context and should name the enrichment source.
Before mutating records, agents should show candidate changes and evidence, then ask for approval.
# Records overview
Source: https://docs.textmine.com/records/overview
Learn how TextMine Records turns document evidence into structured business data with schemas, candidates, review flows, graph views, and enrichment.
Records turn document evidence into structured business data. Use Records when a fact from documents should become a durable business object rather than remain only as a tag, report, or Workbench answer.
![TextMine Records screen showing structured business data generated from document evidence]()
## When to use Records
Use Records for durable entities such as:
* Companies and counterparties.
* Agreements and amendments.
* Vendors and suppliers.
* Customers and accounts.
* People, signatories, and contacts.
* Properties, assets, facilities, or systems.
* Obligations, controls, risks, and issues.
* Custom business objects defined by the team.
Use Vault tags when the fact is mainly document metadata. Use Records when the fact should be reviewed, related to other objects, queried later, enriched, or maintained over time.
## Core concepts
| Concept | Meaning |
| ------------ | ---------------------------------------------------------------------------- |
| Schema | Defines record types, fields, and relationships. |
| Record | A durable object such as a supplier, contract, obligation, or risk. |
| Candidate | A proposed record or field change extracted from documents or tags. |
| Evidence | Source document, tag, passage, table, entity, or file that supports a value. |
| Review queue | Human review of proposed candidates before accepting changes. |
| Graph | Relationship view across records and entities. |
| Enrichment | Supplemental context from configured external or internal sources. |
## How Records relate to Vault
Vault is usually the evidence source. Records use Vault documents, extracted tags, entities, tables, and validated fields to suggest or update structured data.
A common flow is:
1. Upload or import documents into Vault.
2. Process the documents and extract default/custom tags.
3. Extract entities and relationships where useful.
4. Generate record candidates from the evidence.
5. Review and accept candidates.
6. Use Records for graph queries, reporting, workflow routing, and downstream exports.
## Common record workflows
| Workflow | Records role |
| ------------------- | -------------------------------------------------------------------------- |
| Supplier intake | Create supplier, contract, system, and risk records from intake documents. |
| Contract inventory | Turn validated contract metadata into searchable agreement records. |
| Obligation tracking | Extract obligations and link them to parties, dates, and source clauses. |
| Risk register | Create risk or control records from reviewed evidence. |
| Entity graph | Connect companies, people, agreements, assets, and obligations. |
## Agent guidance
Agents should not update Records directly from unvalidated guesses. A good agent flow is to discover schemas, inspect existing records, identify candidate changes, cite source evidence, and ask for approval before accepting or creating durable records.
For API-backed work, use [API and developer docs](/integrations/api-and-developer-docs) and the Public API V3 reference.
# Schemas and review candidates
Source: https://docs.textmine.com/records/schemas-and-candidates
Define record structure and review proposed changes.
Record schemas define the structured data model. Candidates are proposed records or field changes generated from Vault evidence, Workbench activity, workflows, or other configured sources.
## Schemas
A record schema defines:
* Record type name.
* Fields and field types.
* Required and optional fields.
* Relationships to other record types.
* Evidence expectations.
* Review rules and ownership where configured.
Examples:
| Schema | Typical fields | Relationships |
| ---------- | ------------------------------------------------------------------ | --------------------------------------- |
| Supplier | Name, jurisdiction, risk rating, status, contact, source documents | Agreements, systems, controls, issues |
| Agreement | Parties, effective date, expiry, renewal, value, governing law | Counterparties, obligations, amendments |
| Obligation | Owner, due date, trigger, clause source, status | Agreement, party, control |
| Risk | Description, severity, mitigation, owner, review date | Supplier, agreement, control, evidence |
## Candidates
TextMine can suggest candidates from source evidence. A candidate may propose:
* A new record.
* A field value for an existing record.
* A relationship between records.
* A changed value.
* A missing-field recommendation.
* A duplicate or merge candidate.
Candidates should be reviewed before acceptance so records remain trustworthy.
## Candidate review flow
1. Generate or discover candidates from documents, tags, entities, tables, or workflows.
2. Group candidates by schema, record, field, or source document.
3. Inspect the proposed value and confidence where available.
4. Review the source evidence.
5. Accept, reject, edit, or defer the candidate.
6. Record the review outcome and keep the evidence link.
## Evidence
Each record field should remain connected to the documents and extracted evidence that support it. Evidence can include:
* Vault document IDs and names.
* Extracted default or custom tags.
* Entity extraction outputs.
* Table rows or line items.
* Source passages and page numbers.
* Image tags or embedded image regions.
* Workflow run outputs.
* Reviewer decisions.
Use evidence links and audit events to understand where a value came from and when it changed.
## Schema design guidance
Good schemas are specific enough to be useful but not so rigid that every document creates exceptions.
* Start with the business object users already understand.
* Keep fields stable and reviewable.
* Use relationships for connected objects instead of packing everything into one record.
* Prefer controlled fields where reporting depends on consistent values.
* Keep source evidence mandatory for high-risk or regulated fields.
## Agent guidance
Agents should read schemas before creating candidates. They should avoid inventing fields when an existing schema can be used, and should ask before accepting, deleting, or bulk-updating records.
# 15 June 2026
Source: https://docs.textmine.com/release-notes/15-june-2026
Workflow export blocks, Workbench scoping, private folders, and report reliability.
This update improves Workbench source scoping, workflow creation and debugging, report generation, and export paths for operational teams.
## Workbench source scoping
Workbench is better at keeping work tied to the source the user actually asked for. It now preserves scoped requests across follow-up questions, distinguishes uploaded files from Vault documents more reliably, and avoids broadening from a selected folder, Vault, document set, or uploaded source unless the latest request explicitly asks for a broader scope.
This helps tasks such as asking about selected documents, comparing uploaded spreadsheets with Vault evidence, creating agents or workflows for a specific Vault, and turning Workbench analysis into a reusable workflow.
## Workflow builder reliability
The workflow builder now provides stronger validation and clearer warning states while a workflow is being drafted. The data map and canvas highlight missing inputs, incompatible connections, empty-input behavior, and scope issues earlier in the design process.
Workflow run views also show richer execution details, generated artifacts, report outputs, and links back to source Vault documents when available.
## New workflow export and integration options
Workflows can now support more downstream delivery patterns:
* Tableau export blocks for view or workbook outputs, including CSV, Excel, PDF, PNG, and PowerPoint formats.
* Power BI export blocks for report outputs, including PDF, PowerPoint, and PNG formats.
* REST/API request response handling for text, JSON values, and downloadable files.
* ServiceNow submission actions that can attach workflow artifacts, update records, or do both.
* SharePoint submission actions for sending workflow-generated files to configured SharePoint locations.
API and webhook blocks also support secret placeholders in request configuration so teams can keep reusable workflow definitions clean while using protected credentials at run time.
## Better workflow reports
Workflow report generation now carries more document context into report and template steps. This makes scheduled reports, approval packets, agent-generated summaries, and custom report prompts more likely to cite the right source documents and preserve useful artifacts for review.
When a workflow produces files, reports, filled templates, or agent outputs, the run history is better at showing what was generated and where it can be downloaded or followed up.
## Vault folders and table readability
Vault folder privacy controls are more complete. Teams can create or move documents into private folders, preserve parent-folder access behavior, and select the users who should retain access where the folder is private.
Vault table cells and document type labels also render more consistently, especially when extracted values contain highlights or HTML-like source markup.
# 25 June 2026
Source: https://docs.textmine.com/release-notes/25-june-2026
Python workflow blocks, Knowledge Bases, and Playbook analytics.
This release adds more programmable workflow steps, launches Knowledge Bases for shared source libraries and AI-assisted deliverables, and introduces analytics for improving playbooks after real review runs.
## Workflow blocks for data analysis and regulated-source workflows
Workflows now include more blocks for data transformation, analysis, source collection, and review operations.
The biggest addition in [Workflow blocks](/workflows/blocks) is `run_python`. This block runs a bounded Python data-analysis script over workflow inputs, labels, arrays, variables, and table outputs. Workflow labels are exposed as snake\_case variables and `*_list` arrays, while top-level result keys and public variables can be saved back as workflow variables for later blocks.
`run_python` is designed for analysis inside the workflow context: calculations, table cleanup, reconciliation prep, exception summaries, and intermediate values that later reports, records, emails, or approvals can reuse. Network access is blocked in the Python runtime, so external calls should still use `api_request` or `webhook` blocks.
Other new workflow blocks help teams build richer automations without leaving the workflow builder:
* `run_workbench_capability` runs Workbench from a natural-language instruction using the workflow's current Vault, document, and input scope.
* `move_document_to_folder` moves the current workflow document scope into a target Vault folder.
* `resolve_ecfr_sources` fetches official eCFR text from CFR/eCFR citations and can follow capped cross-references.
* `resolve_fca_handbook_sources` fetches FCA Handbook provision text and can follow capped Handbook cross-references.
* `extract_tables_or_line_items` captures tables or line items from in-scope documents for downstream review and reporting.
* `get_analytics_metric` reads workflow analytics, starting with pages processed for an organization or workflow Vault over a selected period.
* `sample_population` selects fixed-size or percentage samples and can emit a sample register for audit and QA.
Together these blocks make workflows better suited to data-heavy processes such as regulatory monitoring, population testing, report generation, control checks, and recurring document reviews.
## Knowledge Bases
[Knowledge Bases](/knowledge-bases/overview) are now available for teams that want a shared, curated place to collect source material and ask questions over it.
A Knowledge Base can hold uploaded files, Vault documents, Legislate documents, and business records. Teams can organize sources into folders, add members, manage private or team visibility, and keep a recent activity trail for the hub.
Each source can be profiled with summaries, document type labels, topics, starter questions, extracted context, tables, entities, and reusable tags. Teams can also define feedback fields so reviewers can capture structured source-level feedback such as status, priority, owner, date, or notes.
Users can ask a Knowledge Base questions and receive answers grounded in the selected hub sources, with source references and evidence where available. Knowledge Bases also support generated artifacts and reports, including document, presentation, PDF, and report-style outputs. Artifacts can be refreshed, revised, versioned, exported to Word, PowerPoint, or PDF, and promoted into a playbook when a drafted review artifact should become reusable review logic.
## Playbook analytics
Playbooks now have a [Playbook analytics](/playbooks/analytics) view that turns completed playbook runs into rule-level feedback.
The analytics dashboard shows how each playbook is performing across completed runs, including compliant rules, non-compliant rules, rules that need review, and rules that were not observed. Teams can filter by playbook, open the related runs, and inspect rule outcomes without leaving the Playbooks area.
For reviewed Word document iterations, analytics can also show resolution signals such as resolved non-compliance, average iterations to resolution, and reviewed versions compared. This helps teams understand whether playbook findings are leading to corrected document language over time.
The new analytics tabs separate consistently compliant rules, rules needing attention, recommendations, potential conflicts, and unobserved rules. Where TextMine detects a recurring issue, analytics can recommend playbook corrections, show the proposed change, and let reviewers apply, discard, restore, or apply all recommendations. Teams can also export playbook analytics to PDF for review packs or governance reporting.
# 7 June 2026
Source: https://docs.textmine.com/release-notes/7-june-2026
Task-agent email activation, operational inboxes, and Copilot Studio setup.
We have made task-agent email easier to use in operational workflows and added guidance for connecting TextMine to Microsoft Copilot Studio.
## Task-agent email for operational requests
Organizations can now route more task-agent work through email once task-agent email has been activated for that organization.
This is useful for shared service accounts, intake inboxes, ticketing systems, Zapier automations, and connected operational channels such as ServiceNow. A team can send a request to its TextMine task-agent address, then continue the same work by replying to the thread.
Task-agent email must be activated before incoming emails can create or update task-agent work. If it has not been activated, TextMine does not create the task and the requester receives a clear message explaining that the email route needs to be enabled first.
## More useful email replies
Task-agent emails now read more like a working conversation.
When TextMine receives a task, the acknowledgement tells the requester that the work has started and that they can reply with extra context. If the task agent needs clarification, the email includes the question and explains that a reply will resume the task.
When a task completes, the result is included directly in the completion email where available, so users do not need to hunt for the outcome in another screen.
## Copilot Studio setup guide
We have added a [Copilot Studio setup guide](/integrations/copilot-studio) for teams that want a Microsoft Copilot Studio agent to use TextMine.
The guide walks through the Azure Marketplace setup, adding TextMine MCP to an agent, and the first checks to run after connecting an agent.
## Document intake reliability
Document intake is more reliable for automated and service-account-driven flows.
This helps requests that arrive from shared inboxes, workflow-generated emails, Zapier automations, ServiceNow-backed processes, and other non-interactive routes.
# Release notes
Source: https://docs.textmine.com/release-notes/index
Current 2026 TextMine product release notes.
Release notes are limited to 2026 for now. Older imported product updates remain available in the [Intercom archive](/archive/intercom).
* [25 June 2026](/release-notes/25-june-2026)
* [15 June 2026](/release-notes/15-june-2026)
* [7 June 2026](/release-notes/7-june-2026)
* [26 May 2026](/archive/intercom/product-updates/26-may-2026)
* [8 May 2026](/archive/intercom/product-updates/8-may-2026)
* [25 March 2026](/archive/intercom/product-updates/25-march-2026)
* [12 March 2026](/archive/intercom/product-updates/12-march-2025)
* [5 March 2026](/archive/intercom/product-updates/5-march-2025)
* [19 February 2026](/archive/intercom/product-updates/19-february-2026)
* [12 February 2026](/archive/intercom/product-updates/12-february-2026)
* [28 January 2026](/archive/intercom/product-updates/28-january-2026)
* [19 January 2026](/archive/intercom/product-updates/19-january-2026)
# Reports and exports
Source: https://docs.textmine.com/reports-and-exports/overview
Generate, review, automate, and export TextMine outputs with source evidence.
Reports and exports turn TextMine analysis into shareable outputs. They can be generated from Vault, Workbench, Workflows, Playbooks, Records, and API-backed processes.
Use reports when a user needs a narrative, table, review memo, exception list, or evidence-backed summary. Use exports when the output should move into another system, spreadsheet, archive, or workflow step.
## Where reports come from
| Source | Common report type | Notes |
| ----------------- | ------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------- |
| Vault / Ask Vault | Document summaries, comparison reports, extracted-field reports, table summaries | Best when the report is grounded in selected Vault documents. |
| Workbench | Exploratory research outputs, multi-file analysis, generated tables, documents, charts, and follow-up files | Best for supervised, iterative work. |
| Workflows | Scheduled reports, approval packets, exception reports, template-based reports, BI exports, integration submissions | Best for repeatable processes and audit history. |
| Playbooks | Clause review outputs, policy checklists, negotiation guidance, comparison artifacts | Best for reusable review rules. |
| Records | Entity summaries, missing-field reports, graph or relationship summaries, candidate review reports | Best for structured business data. |
| Public API / MCP | Programmatic report generation, status polling, and workflow-backed file outputs | Best for developer and agent workflows. |
## Report types
TextMine can support several output shapes depending on the originating surface and configuration:
* Free-form narrative reports.
* Structured tables and CSV-style outputs.
* Word or document-style review artifacts.
* Spreadsheet or template-filled outputs.
* Exception registers.
* Approval packets.
* Evidence-backed summaries with source documents and passages.
* Export files for external systems.
* Tableau and Power BI export artifacts.
## Workflow report blocks
Workflows can generate reports as part of a repeatable process.
| Block | Use it for |
| ------------------------------------ | -------------------------------------------------------------------------------------------------- |
| `send_report_template` | A standard report template exists and should be reused with the current workflow document context. |
| `send_report_custom` | The report should be generated from a prompt or custom instruction grounded in workflow scope. |
| `fill_template` | A spreadsheet, CSV, text, Word, or form template should be filled from workflow values. |
| `api_request` | A REST call should return text, extracted JSON values, or a downloadable file for later steps. |
| `tableau_export` / `power_bi_export` | BI content should be exported as a workflow artifact. |
| `send_email` | The generated report or export should be delivered to recipients. |
| `submit_to_integration` | The output should be sent to an external system such as SharePoint or ServiceNow where enabled. |
## Exports and external systems
Exports can be used to move TextMine outputs into downstream systems. Depending on configuration, this may include:
* Downloading generated files from Workbench or report outputs.
* Sending workflow outputs by email.
* Submitting workflow outputs to SharePoint.
* Attaching workflow artifacts to ServiceNow records or updating ServiceNow fields.
* Publishing validated documents from Vault to Legislate.
* Exporting Records data for review or external sync.
* Using Public API V3 to retrieve generated outputs or poll asynchronous jobs.
Use [API and developer docs](/integrations/api-and-developer-docs) when an agent or developer needs exact endpoints, operation IDs, schemas, or MCP setup.
## Evidence and audit guidance
High-quality reports should include enough context for a reviewer to trust and reproduce the result:
* Source document names and IDs.
* Vault, folder, workflow, report, or record IDs where available.
* Extracted tags, tables, entities, or record fields used.
* Source passages, table references, image regions, or page numbers where available.
* Processing status and validation state.
* Reviewer or approver decisions when applicable.
Agents should avoid presenting generated reports as validated evidence unless the report includes source references or a human has reviewed the output.
## When to automate a report
Automate a report with Workflows when it should run repeatedly, on a schedule, after import, after processing, or after approval. Keep the report in Workbench when the user is still exploring the question and changing the shape of the output.
Good automation candidates include:
* Weekly renewal or expiry reports.
* Procurement intake summaries.
* Legal playbook review packets.
* Risk exception registers.
* SEC filing monitor summaries.
* Records candidate review reports.
* Table extraction QA reports.
## Agent guidance
Before generating or exporting a report, agents should confirm:
1. Which documents, records, workflow run, or provider import are in scope.
2. Whether the output is exploratory or final.
3. Whether a template should be used.
4. Whether the report should include source evidence.
5. Who should receive the output.
6. Whether the output should be saved, emailed, exported, submitted to an integration, or published to Legislate.
# Scribe overview
Source: https://docs.textmine.com/scribe/overview
Create and fill smart document templates.
Scribe is the template and document-generation surface for TextMine. Use it to import Word templates, define fields, add conditional logic, publish templates to teams, and generate documents from structured answers or TextMine data.
## What Scribe is for
Scribe helps teams create repeatable documents from structured inputs. Typical uses include:
* Supplier onboarding forms.
* Contract templates.
* Review memos.
* Internal approval documents.
* Engagement letters.
* Policy documents.
* Generated summaries or packs from Workbench and Workflows.
## Template building blocks
| Building block | What it does |
| ----------------- | ----------------------------------------------------------------------------------------------- |
| Template | The reusable document or form structure. |
| Field | A variable value that can be filled by a user, workflow, Workbench, or extracted data. |
| Field group | A logical set of related fields. |
| Conditional logic | Shows, hides, or changes content based on answers or values. |
| Word import | Converts a Word document into a Scribe template while preserving as much structure as possible. |
| Team publishing | Makes a template available to selected teams or users. |
## Word template import
Use Word import when a team already has a document template. After import, review the converted structure, map placeholders to Scribe fields, define field groups, and test generated output before publishing.
Imported templates should be checked for:
* Repeated clauses or sections.
* Tables and numbering.
* Conditional language.
* Placeholder names.
* Required and optional fields.
* Team visibility.
* Generated document formatting.
## Filling templates
Templates can be filled manually or as part of broader TextMine workflows. Values may come from:
* User answers.
* Workbench context and generated values.
* Workflow outputs.
* Vault tags and extracted fields.
* Records data.
* Uploaded or imported documents.
Workbench can help infer fill values from context. Workflows can fill spreadsheet, CSV, text, Word, or form templates and use the generated file in later steps, reports, emails, approvals, or exports.
## When to use Scribe
Use Scribe when the desired output is a generated document or template. Use Reports when the output is primarily analysis or evidence. Use Workflows when template generation should be triggered, approved, emailed, or integrated with another system.
## Agent guidance
Agents should inspect available templates and fields before proposing new ones. When creating or updating a template, return the intended fields, field groups, conditional logic, source data, and publication scope. Ask for confirmation before publishing a template to a team or using a generated document externally.
# Contact support
Source: https://docs.textmine.com/support
How to contact TextMine support and what to include so the team can route product, access, integration, billing, or workflow issues quickly.
Use this page when the Help Center does not answer your question, you need a tenant or feature enabled, or an issue needs TextMine support to investigate.
## Email support
Email [support@textmine.com](mailto:support@textmine.com) for TextMine product, access, configuration, integration, billing, or workflow support.
Use your organization's approved support route if your TextMine deployment has a dedicated customer support process or named TextMine contact.
## What to include
Include enough context for the team to route the request without extra back-and-forth:
* Your organization or tenant name.
* The TextMine area involved, such as Workbench, Vault, Workflows, Records, Integrations, Public API, MCP, or Scribe.
* The page URL, Vault, workflow, record, document type, run, or request ID where relevant.
* What you expected to happen and what happened instead.
* Steps to reproduce the issue, including screenshots when useful.
* Business impact, urgency, and any deadline.
* Whether the issue affects one user, a team, or the whole organization.
Do not send passwords, API keys, authentication tokens, private keys, or other secrets by email. If support needs sensitive configuration, ask for the approved secure handoff route.
## Common support requests
| Request | Useful details |
| ---------------------------------- | ----------------------------------------------------------------------------------------- |
| Access or permissions | User email, tenant, team, role needed, and the business reason. |
| Feature activation | Tenant, feature, deployment type, and who approved the change. |
| Workflow or task issue | Workflow name, run ID, trigger type, inputs, expected output, and error message. |
| Vault or extraction issue | Vault name, document type, affected document ID, tags or fields, and expected extraction. |
| Integration or API issue | Provider, endpoint or tool, request ID, timestamp, status code, and sanitized error text. |
| Billing or third-party agent setup | Billing contact email, organization name, signup route, and provisioning status. |
## Before sending
Search the Help Center first if the question is about a known workflow, integration, permission, or release note. Good starting points:
* [Quick start](/getting-started/quick-start)
* [Product map](/getting-started/product-map)
* [Permissions](/administration/permissions)
* [Integrations](/integrations/overview)
* [Workflows](/workflows/overview)
* [Release notes](/release-notes)
If you are reporting a docs problem rather than a product issue, use the **Raise issue** link at the bottom of the affected page.
# Ask Vault and reports
Source: https://docs.textmine.com/vault/ask-vault-and-reports
Ask questions and generate reports from Vault evidence.
Ask Vault answers questions using Vault documents, extracted tags, and optional uploaded reference text.
For the broader Vault capability map, see [Vault and editor capabilities](/vault/capabilities).
## Ask a question
Use Ask Vault for quick answers, comparisons, summaries, or extraction questions. If the question names a specific folder, subset, or document group, resolve those documents first rather than asking the whole Vault.
## Create reports
Use reports when you need a saved analysis, table, comparison, or evidence-backed output. Reports can be created from selected documents, search results, or resolved document groups.
## Sources
Good Vault answers should stay close to the source documents and extracted evidence. For important outputs, include the source document names and the fields or text used to reach the answer.
# Vault and editor capabilities
Source: https://docs.textmine.com/vault/capabilities
Reference for Vault document management, extraction, Ask Vault, and the validation editor.
Vault is the governed document intelligence surface in TextMine. It stores source documents, runs processing and extraction, supports review in the Vault editor, and feeds Workbench, Workflows, Playbooks, Records, Agents, Reports, and Legislate.
Use this page as a functional map for Vault and the Vault editor.
## Vault surfaces
| Surface | What it is for |
| ------------------- | ----------------------------------------------------------------------------------------------------------------- |
| Vault document list | Browse, search, filter, sort, assign, archive, and open documents in a Vault or folder. |
| Document types | Configure the categories and default fields used for extraction and review. |
| Custom tags | Create and run reusable questions, chained tags, entity extraction tags, table-aware tags, and image tags. |
| Ask Vault | Ask questions, create reports, run Vault-scoped agents/workflows, and inspect long-running outputs. |
| Vault editor | Review a single processed document, validate extracted values, inspect sources, tables, images, and entity graph. |
| Pop-out views | Open tag cards, extracted tables, or entity graph in focused full-window views. |
## Document management
Vault can manage documents through the full document lifecycle:
* Upload or import files into a Vault.
* Organize documents into folders.
* Browse document tables with configurable columns.
* Search, filter, and sort by document name, type, folder, status, assignee, and extracted values where available.
* Create or move documents into private folders and select which users should retain access.
* Select documents for batch actions.
* Assign or unassign reviewers and assignees.
* Archive, restore, or remove documents where permissions allow.
* Open documents in the validation editor.
* Track processing, publication, failure, and completion states.
Agents should distinguish document identifiers carefully. A Vault document often has both a document/Vault ID and a contract-configuration-style ID used by older app surfaces.
## Document types and default tags
Document types define the extraction profile for a class of documents. They can include default tags, prompts, validation expectations, and Vault-specific processing settings.
Vault settings support:
* Creating document types.
* Updating document type settings.
* Deleting custom document types when not in use and permitted.
* Assigning document types to Vaults.
* Copying or reusing default tags.
* Promoting or sharing document type configuration where supported.
* Enabling Vault-specific processing settings, such as table extraction, embedded image classification, extensive text extraction, and image model processing.
## Custom tags and questions
Custom tags let a team ask repeatable questions over documents and persist the answers. See [Tags and extraction](/vault/tags-and-extraction) for the detailed reference.
Supported patterns include:
* Fact-style questions over document text.
* Aggregation tags for broader retrieval and synthesis.
* Chained tags that use custom tags, default tags, entities, or tables as context.
* Entity resolution tags that extract named entities and descriptions.
* Image tags that use selected screenshots or embedded images.
* Tags with reasoning or semantic card output.
* Bulk tag runs over selected documents or whole Vault scopes.
After creating or changing tags, selected documents may need to be processed or reprocessed before answers are current.
## Extraction and processing
Vault processing can include:
* Document type prediction or assignment.
* Text extraction and OCR where applicable.
* Default tag extraction.
* Custom tag/question extraction.
* Embedded image extraction and classification.
* Image transcription through image types and image tags.
* Table extraction and line-item extraction.
* Entity extraction and entity relationship resolution.
* Document reprocessing after configuration changes.
Long-running extraction should expose status to users or agents. Workbench and workflows should wait for processing before using extracted values.
## Ask Vault
Ask Vault answers questions using Vault documents, extracted tags, selected documents, and optional uploaded reference files.
Ask Vault can support:
* Quick answers with source documents.
* Comparisons between documents or uploaded reference material.
* Saved report generation.
* Tables and chart-ready summaries.
* Running on-demand agents over Vault documents.
* Running workflows from a Vault context.
* Tracking report, tag, document-processing, workflow, and agent status.
Good answers should include the source document names, IDs, and fields or passages used.
## Vault editor
The Vault editor is the single-document review and validation surface. It is used after a document has been uploaded or imported and processing has begun or completed.
The editor supports:
* Viewing the document with pagination and search.
* Reviewing extracted default tags and custom tags in cards.
* Editing extracted values where permitted.
* Validating or correcting values against source evidence.
* Seeing source passages and highlights linked to fields.
* Sorting and organizing field cards.
* Opening tag cards in a pop-out view.
* Viewing extracted tables and line items.
* Opening tables in a dedicated pop-out table view.
* Viewing resolved entities and relationships in an entity graph.
* Opening the graph in a pop-out view with zoom, filtering, and PNG export.
* Creating new custom questions from the document context.
* Creating image tags from selected document screenshots.
* Viewing uploaded Word or PDF-style documents through the appropriate visualizer.
The editor should be used when a human needs to inspect evidence, correct values, approve extraction, or understand how the system reached an answer.
## Tables, entities, and images
Vault-specific analysis features include:
| Feature | What it does |
| -------------------- | -------------------------------------------------------------------------------------- |
| Table extraction | Detects and extracts tables or line items from in-scope documents. |
| Table editor/pop-out | Lets users inspect extracted tables in a focused view. |
| Entity extraction | Extracts entities such as parties, people, organisations, or domain-specific entities. |
| Entity graph | Displays entity relationships and evidence, with pop-out graph controls. |
| Screenshot selector | Lets users select image regions from a document for image-based tags. |
| Image type settings | Configure image transcription/classification behavior for a Vault/document type. |
## Publishing to Legislate
Vault is the source and review layer. Legislate is the validated archive. When documents or data are ready, TextMine can publish from Vault to Legislate so the validated version appears in the Legislate team-document surface.
Agents should not treat a Vault document as published just because it exists in Vault. Look for the document state or the Legislate record when the question is about published data.
## Vault integrations
Vault can receive documents from uploads, Workbench imports, connected providers, native integration flows, and workflows. See [Integration catalog](/integrations/catalog) for the provider list, including ServiceNow, SharePoint, DocuSign, Google Drive, and MCP/provider scan-import options.
# Document types and tags
Source: https://docs.textmine.com/vault/document-types-and-tags
Configure extraction fields for Vault documents.
Document types and tags define what Vault should extract and review. For the full tag and extraction reference, see [Tags and extraction](/vault/tags-and-extraction). For the broader Vault editor reference, see [Vault and editor capabilities](/vault/capabilities).
## Document types
Use document types to group documents with similar extraction needs, such as NDAs, service agreements, invoices, purchase orders, utility bills, SEC filings, or custom business documents.
A document type can define the expected default tags for that class of document. It can also be associated with Vault-specific processing settings such as table extraction, embedded image classification, extensive text extraction, and image model processing.
## Default tags
Default tags are reusable extraction fields for a document type. They are best for stable fields that should be extracted whenever a document of that type is processed.
Examples include parties, effective date, governing law, renewal date, termination notice period, contract value, supplier name, clause presence, and other recurring metadata.
Default tags are part of the document type configuration. If default tags change, reprocess affected documents when you need updated answers.
## Custom tags
Custom tags are user-defined questions that can be created and run against selected documents or Vault scopes. They are useful when a team needs organisation-specific extraction beyond the default document type fields.
Custom tag modes include:
| Tag type | Use it for |
| ---------------- | -------------------------------------------------------------------- |
| Fact-style tags | Specific values or answers grounded in source passages. |
| Aggregation tags | Broader synthesis across multiple passages or sections. |
| Chained tags | Questions that depend on other tags, entities, or tables. |
| Entity tags | Extracting named entities and descriptions for graph or record use. |
| Image tags | Asking multimodal questions over selected or detected image regions. |
Custom tags can also include reasoning, semantic-card output, validation state, source passages, page references, and processing status.
## Chained tags, entities, and tables
Chained tags let a later question use earlier extraction as context. Sources can include custom tags, default tags, extracted entities, and extracted tables.
Use chained tags when the answer requires intermediate structure. For example, first extract parties, then ask which party has each termination right; or extract a pricing table, then ask for the highest-risk pricing obligation.
Entity tags are designed for people, organisations, parties, assets, obligations, or other named concepts that should feed records, graphs, or relationship-aware workflows.
Table extraction detects tables and line items so they can be reviewed in the Vault editor, used in reports, or passed into chained tags and workflows.
## Image and embedded-image extraction
Vault can support image transcription/classification, embedded image extraction, table extraction, and line item extraction where configured.
Image tags use multimodal question answering over image regions. Embedded image detection can render document pages, classify visual regions against configured image types, and run image-type questions on the detected image context.
Use this for ownership charts, signatures, scanned schedules, diagrams, and visual tables that are not reliably represented as ordinary text.
## Reprocessing
When document configuration changes, selected documents can be reprocessed so extracted values reflect the latest tags, document type configuration, or processing pipeline.
Reprocess after changing default tags, extraction settings, image settings, table settings, or upstream sources used by chained tags.
# Vault overview
Source: https://docs.textmine.com/vault/overview
Learn how TextMine Vault stores, classifies, extracts, searches, validates, reports on, and publishes document intelligence.
Vault is the document intelligence surface in TextMine. Use it to collect documents, organize them in folders, classify document types, extract default and custom tags, ask questions, run reports, and feed downstream workflows, playbooks, agents, and records.
For the full Vault and validation-editor reference, see [Vault and editor capabilities](/vault/capabilities).
## Core concepts
* A Vault contains documents and folders for a team or business unit.
* Folders can be public to the team or private to selected users, depending on permissions.
* Document types determine the default extraction and review fields for each document.
* Custom tags and questions capture organization-specific data points.
* Ask Vault can answer from selected documents, all accessible Vault documents, extracted tags, and uploaded reference material.
* Reports, agents, playbooks, and workflows can all run against Vault documents.
* Validated Vault documents can be published into Legislate, where they become part of the team-document archive.
## Related features
Use Workflows to automate document events. Use Playbooks to apply reusable review logic. Use Records to turn evidence into structured data. Use Legislate for the published team-document view. Use Workbench when the task needs cross-product orchestration.
# Tags and extraction
Source: https://docs.textmine.com/vault/tags-and-extraction
Default tags, custom tags, chained tags, entity tags, image tags, tables, and extraction settings.
Tags define what TextMine should extract, review, and reuse from Vault documents. They can be document-type defaults, one-off document questions, reusable custom tags, chained outputs, entity extraction prompts, image tags, or table-aware questions.
Use this page when designing extraction configuration or explaining how Vault tags work.
## Tag scopes
| Scope | What it means | Common use |
| -------------------------- | ------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| Document type default tags | Tags attached to a document type and applied to matching documents. | Standard fields such as parties, dates, governing law, value, renewal terms, or document-specific metadata. |
| Vault custom tags | Reusable questions configured for a Vault, document type, or selected documents. | Organisation-specific extraction and review fields. |
| Document-level tags | Tags or questions created from the validation editor for a specific document context. | Ad hoc review, follow-up extraction, or evidence correction. |
| Image type tags | Tags configured for image types and applied to detected or selected image regions. | Ownership charts, signatures, diagrams, tables captured as images, or other embedded visual content. |
A tag result can include a value, reasoning, source passages, preview offsets, page number, validation state, status, model metadata, and whether the tag came from a default configuration.
## Default tags
Default tags are reusable extraction fields for a document type. They are the best fit when the same field should be extracted every time a document of that type is processed.
Use default tags for stable fields such as:
* Effective date.
* Counterparty.
* Governing law.
* Termination notice period.
* Contract value.
* Renewal date.
* Supplier name.
* Clause presence or absence.
Default tags can be copied, reused, and applied when documents are processed or reprocessed. If a document type or its default tags change, affected documents may need reprocessing before results reflect the new configuration.
## Custom tags
Custom tags are user-defined questions that can be run against selected documents or a Vault scope. They are useful when the field is not part of the standard document type or when a team needs temporary or evolving extraction.
Common custom tag modes include:
| Mode | Internal analysis type | Use it when |
| --------------- | ---------------------- | --------------------------------------------------------------------------------------- |
| Fact-style tag | `FACT_FLOW_RAG` | The answer should come from a specific fact or passage in the document. |
| Aggregation tag | `DEEP_RAG` | The answer needs broader retrieval, synthesis, or aggregation across multiple passages. |
| Entity tag | `ENTITY_RESOLUTION` | The output should be a list of entities with optional descriptions. |
| Image tag | `MULTI_MODAL_QA` | The answer should come from a selected image or detected embedded image. |
Custom tags can also request reasoning, produce semantic-card style output, or be chained from earlier extraction context, depending on configuration.
## Aggregation tags
Aggregation tags use a deeper retrieval and reasoning path. Use them when a simple field extraction is too narrow.
Good examples:
* Summarise all payment obligations in this agreement.
* Identify every termination right and group by party.
* Compare all references to change-of-control restrictions.
* Aggregate all dates that affect renewal, notice, or expiry.
Avoid aggregation tags when the desired answer is a single obvious field. A default or fact-style tag is usually easier to validate and rerun.
## Chained tags
Chained tags use earlier context as inputs for a new question. A chained tag can reference:
* Other custom tags.
* Default tags.
* Extracted entities.
* Extracted tables.
Use chained tags when the answer depends on intermediate extraction. For example:
* Use extracted parties and governing law to assess jurisdiction-specific risk.
* Use table extraction as context for a covenant or pricing answer.
* Use entity extraction plus contract dates to build a party-obligation summary.
* Use a default tag value as a filter or comparison point for a custom tag.
Chained tags cannot be treated like independent extraction. Agents should inspect the source tags first, confirm their status, and rerun upstream tags or tables if the context is stale.
## Entity tags
Entity tags use entity resolution. Instead of a free-form question, the user defines entity names or categories, optionally with descriptions. TextMine extracts and resolves those entities from the document context.
Use entity tags for:
* Parties and affiliates.
* People, roles, signatories, or contacts.
* Products, assets, facilities, suppliers, or properties.
* Legal concepts that should become nodes in a graph.
* Domain-specific entities that Records or Workbench should use later.
Entity extraction can feed the Vault entity graph, Records, workflows, and chained tags. Include descriptions when entity names are ambiguous.
## Image tags
Image tags use multimodal question answering over an image region. They can be created from a selected screenshot region in the Vault editor or from embedded image detection.
Image tags store image context such as page number, coordinates, width, height, and the uploaded image asset. They are useful for:
* Ownership charts.
* Signature blocks.
* Scanned tables.
* Organograms and diagrams.
* Embedded images that contain legally or commercially meaningful text.
Image tags are different from text tags. Aggregation and semantic-card modes are disabled for image tags because the source is the selected image context rather than text retrieval over the full document.
## Embedded image detection
Embedded image detection is a Vault/document-type processing setting. When enabled, TextMine can render document pages, classify visual regions against configured image types, and apply image-type questions to detected image content.
The flow is:
1. Configure image types and image tags for the Vault or document type.
2. Enable embedded image classification/image processing where appropriate.
3. Process or reprocess documents.
4. TextMine detects matching embedded image regions.
5. Matching image tags run as multimodal questions against the detected image context.
6. Users review the image-tag answers in the editor like other extracted values.
Use embedded image detection when important evidence is visually embedded rather than available as ordinary text. It is especially useful for charts, scanned excerpts, diagrams, and visual schedules.
## Table extraction
Table extraction detects tables and line items from documents so they can be reviewed, exported, used by reports, or used as context for chained tags and workflows.
Use table extraction for:
* Pricing schedules.
* Fee tables.
* Covenant calculations.
* Line-item invoices.
* Obligation matrices.
* Disclosure schedules.
* SEC filing tables.
Tables can be viewed in the Vault editor and in a pop-out table view. Workbench and workflows can also queue table extraction and wait for completion before using table outputs.
## Extraction settings
Vault/document-type settings can control several processing behaviors:
| Setting | What it affects |
| ----------------------------- | -------------------------------------------------------------------------------------- |
| Table extraction | Whether tables and line items should be detected and extracted. |
| Embedded image classification | Whether document pages should be scanned for configured image types. |
| Extensive text extraction | Whether TextMine should use a broader text extraction path for documents that need it. |
| Image model processing | Whether image-oriented model processing should be used for image/document contexts. |
These settings may be inherited from Vault defaults or overridden at a document-type/Vault association. Reprocess documents after changing settings if existing results need to be refreshed.
## Status and validation
Tag answers move through processing states and may need human validation. A good review should check:
* The extracted value.
* The source passage, table, entity, or image region.
* The page and preview offsets where available.
* Whether the tag is current with the latest document type/settings.
* Whether upstream chained sources completed successfully.
For durable downstream use, prefer validated tags or outputs with clear source evidence.
## Agent guidance
Agents should not create tags blindly. Before creating or running tags, they should:
1. Read existing document types and tags.
2. Prefer default tags for stable document-type fields.
3. Use custom tags for team-specific or evolving questions.
4. Use aggregation tags only when synthesis is needed.
5. Use chained tags only after checking upstream tags, entities, or tables.
6. Use entity tags when output should feed graphs, records, or relationship-aware workflows.
7. Use image tags for visual evidence.
8. Ask for approval before creating tags, running tags across many documents, changing extraction settings, or reprocessing documents.
# Workbench capabilities
Source: https://docs.textmine.com/workbench/capabilities
Reference for what Workbench can read, create, run, import, and verify.
Workbench is an in-app AI workspace with a backend capability registry. Each capability has a purpose, expected inputs, outputs, approval rules, and sometimes an asynchronous wait step.
Use this page to understand what Workbench can do. External agents should use [TextMine MCP server](/integrations/textmine-mcp-server) or [API and developer docs](/integrations/api-and-developer-docs) when they need direct API or MCP tool access.
## Capability rules
* Read capabilities can usually run without approval.
* Mutating configuration or durable data requires approval.
* Long-running actions have matching wait capabilities.
* Workbench should inspect or search documents before acting on a vague source scope.
* Saving files into Vault is different from using files as temporary Workbench context.
## Document and Vault capabilities
| Capability | What it does | Notes |
| ----------------------------------- | ------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------- |
| `inspect_vault_documents` | Reads document metadata, IDs, Vault scope, and processing state. | Use early when a task depends on Vault documents. |
| `search_accessible_vault_documents` | Searches documents and extracted tags across accessible Vaults. | Resolves source scopes before asking Vault, creating reports, or running workflow-backed actions. |
| `ask_vault` | Answers using Vault documents, extracted tags, and optional uploaded reference text. | Read-only analysis, comparison, summarization, and evidence-backed answers. |
| `answer_reference_document` | Answers directly from uploaded reference text without requiring Vault processing. | Good for quick uploaded-file analysis. |
| `import_uploaded_documents` | Imports Workbench-uploaded files into a selected Vault. | Requires approval and uses the durable Vault document pipeline. |
| `reprocess_documents` | Queues selected Vault documents for reprocessing. | Requires approval; use after configuration or extraction changes. |
| `wait_for_document_processing` | Waits for document or tag processing to finish. | Used after custom tag runs, imports, or reprocessing. |
## Legislate and comparison capabilities
| Capability | What it does | Notes |
| ----------------------------------- | -------------------------------------------------- | ------------------------------------------------------------------------------- |
| `read_legislate_documents` | Reads active Legislate/team documents. | Legislate is the validated team-document archive, not a Vault. |
| `compare_legislate_vault_documents` | Compares Legislate documents with Vault documents. | Use when the user asks about published documents versus Vault source documents. |
## MCP provider capabilities
| Capability | What it does | Notes |
| ---------------------- | ---------------------------------------------------------------- | --------------------------------------------------------------------------------- |
| `list_mcp_providers` | Lists enabled MCP providers for the Workbench session. | Shows connection status, Workbench enablement, and capabilities. |
| `list_mcp_folders` | Lists folders from a connected file-storage provider. | Requires a provider with folder listing capability. |
| `list_mcp_files` | Lists or searches files in a provider folder. | Use before selecting files for import. |
| `import_mcp_documents` | Imports selected provider files into the current Workbench chat. | Imports temporary Workbench context; saving to Vault is a separate approved step. |
The provider import path is intentionally bounded: Workbench imports selected files, not entire provider repositories. The backend limits a single import batch to selected files and a total size cap.
## Playbooks
| Capability | What it does | Notes |
| ----------------- | ----------------------------------------------------------- | -------------------------------------------- |
| `read_playbooks` | Lists or searches available playbooks. | Read-only inventory. |
| `create_playbook` | Creates a reusable playbook from source text or rules. | Requires approval. |
| `run_playbook` | Runs an existing playbook against selected Vault documents. | Asynchronous; use for reusable review logic. |
## Document types, tags, and extraction setup
| Capability | What it does | Notes |
| --------------------------- | ---------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| `read_document_types` | Lists document types for a Vault or workspace. | Read-only inventory. |
| `create_document_type` | Creates a document type or adds default tag definitions. | Requires approval. |
| `delete_document_type` | Deletes a custom document type. | Requires approval. |
| `setup_image_transcription` | Creates or reuses an image type and default image tags. | Requires approval; used for embedded image transcription/classification. |
| `create_custom_tags` | Creates or reuses custom tags/questions. | Requires approval because it changes reusable extraction configuration. See [Tags and extraction](/vault/tags-and-extraction). |
| `run_custom_tags` | Runs tags/questions against selected documents or a Vault. | Requires approval; often followed by `wait_for_document_processing`. |
| `run_table_extraction` | Queues table or line-item extraction. | Asynchronous; followed by `wait_for_table_extraction`. |
| `run_entity_extraction` | Queues entity extraction/resolution. | Asynchronous; followed by `wait_for_entity_extraction`. |
| `read_vault_entity_graph` | Reads entities and relations from a Vault. | Read-only graph data. |
## Reports and templates
| Capability | What it does | Notes |
| ----------------------------- | ------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- |
| `create_report` | Creates a report, table, comparison, summary, or analysis over selected documents. | Asynchronous; should honor resolved source scope and carry document context into the report. |
| `wait_for_report_completion` | Waits for a report job to complete. | Reads status and result. |
| `fill_template` | Fills an uploaded Word, Excel, CSV, text, form, or spreadsheet template. | Uses uploaded template files and conversation context. |
| `synthesize_workbench_result` | Produces a partial answer from successful prior steps if a dependent analyzer failed. | Fallback for resilient task completion. |
## Workflows and agents
| Capability | What it does | Notes |
| -------------------------- | ------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `read_workflows` | Lists available workflows. | Read-only inventory. |
| `create_workflow` | Creates a reusable workflow using the workflow block catalog. | Requires approval; should preserve the selected Vault or document scope and surface warnings before creation. See [Workflow blocks](/workflows/blocks). |
| `run_workflow` | Runs an existing workflow or one-off workflow-backed action. | Asynchronous; followed by `wait_for_workflow_completion`; should confirm scope before side-effecting runs. |
| `cleanup_one_off_workflow` | Soft-deletes a temporary one-off workflow after completion. | Host-inserted cleanup. |
| `read_agents` | Lists on-demand agents. | Read-only inventory. |
| `create_agent` | Creates an on-demand agent. | Requires approval. |
| `run_agent` | Runs an existing on-demand agent against selected documents. | Requires approval; followed by `wait_for_agent_completion`. |
## Records and schemas
| Capability | What it does | Notes |
| ------------------------ | ------------------------------------ | --------------------------- |
| `inspect_record_schemas` | Reads accessible record schemas. | Use before record mutation. |
| `lookup_records` | Reads or searches business records. | Read-only. |
| `upsert_record` | Creates or updates business records. | Requires approval. |
| `update_record_schema` | Creates or replaces record schemas. | Requires approval. |
## Workbench skills and task lifecycle
| Capability | What it does | Notes |
| ------------------------- | ------------------------------------------------------------------------ | -------------------------------------------------------------------- |
| `read_workbench_skills` | Reads saved Workbench skills from task context. | Use when the user references recurring instructions or review style. |
| `apply_workbench_skill` | Applies saved skill guidance to planning and execution. | Guidance-only capability. |
| `verify_goal_completion` | Verifies that completed task steps satisfy the original goal. | Used near the end of a plan. |
| `save_workbench_artifact` | Saves a result, report, table, or structured output back to the session. | Final persistence step. |
| `ask_for_approval` | Pauses for user approval before a proposed action. | Reserved for host-created approval steps. |
## Safe-use guidance for agents
Agents should return the capability names they used, the IDs they touched, and any approvals they requested. For durable mutations, include enough context for audit: Vault IDs, document IDs, workflow IDs, run IDs, record IDs, report IDs, provider IDs, and imported file names.
# Goals, tasks, and artifacts
Source: https://docs.textmine.com/workbench/goals-tasks-and-artifacts
How Workbench handles longer-running work.
Workbench can turn a request into a task plan, execute capability steps, wait for long-running operations, verify the result, and save a final artifact.
For the full capability list, see [Workbench capabilities](/workbench/capabilities).
## Tasks
Tasks are used when a request needs backend execution, asynchronous processing, source-scope checks, or approval. Examples include running table extraction, generating reports, running agents, applying playbooks, importing documents, creating workflows, or updating records.
Workbench should resolve the intended source before starting a task. That source may be selected Vault documents, a folder, a Vault, uploaded files, a provider import, a workflow run, a record set, or a previous artifact.
## Approvals
Some Workbench capabilities mutate reusable data or trigger broad processing. These actions require approval, such as creating workflows, creating playbooks, importing documents into Vault, creating tags, running agents, updating records, or changing record schemas.
Approval prompts should show the important scope and side effects: which Vault, documents, folders, workflow, records, provider files, or output destinations are involved.
## Artifacts
When a task produces a useful answer, report, table, file, or structured output, Workbench can save it back to the session as an artifact so it can be reviewed later.
Artifacts can also serve as source context for the next step. For example, a generated table can feed a report, a filled template can be attached to an email or integration step, and a workflow draft can be reviewed before it is created.
# Workbench overview
Source: https://docs.textmine.com/workbench/overview
Learn how to use TextMine Workbench to ask questions, inspect Vault documents, import files, create reports, run agents, and save reusable artifacts.
Workbench is the AI workspace for getting work done across TextMine. It can answer questions, inspect Vault documents, import files, create reports, run agents, create and run workflows, apply playbooks, update records, fill templates, and save reusable artifacts.
For the full backend capability reference, see [Workbench capabilities](/workbench/capabilities).
![TextMine Workbench screen showing an AI workspace for scoped document questions and reusable outputs]()
## What Workbench can do
* Inspect selected Vault documents before acting.
* Search accessible Vaults and return document or tag tables.
* Keep follow-up work scoped to the selected Vault, folder, document set, uploaded file, provider import, or previous result unless the user asks to broaden it.
* Read Legislate team-document inventory when the user asks about published Legislate documents.
* Ask Vault questions using documents, extracted tags, and uploaded reference files.
* Analyze uploaded files as temporary context without requiring Vault processing.
* Import uploaded files into a Vault when approved.
* Browse and import files from enabled MCP providers such as connected storage systems.
* Create reports, workflows, playbooks, document types, tags, agents, records, and record schemas.
* Run existing workflows, playbooks, agents, table extraction, entity extraction, and document reprocessing.
* Save final answers as Workbench artifacts.
## When to use Workbench
Use Workbench when the answer is not just "open this page and click a button". It is best for multi-step work, exploratory analysis, document comparison, one-off report generation, and drafting reusable automation.
For straightforward document review, start in Vault. For a repeatable process, create a Workflow. For reusable review rules, create a Playbook.
## Workbench and agents
Workbench is the in-app agentic workspace. It is the right place for a human to supervise analysis, review generated outputs, approve Vault changes, and turn one-off work into reusable workflows or playbooks.
External agents should not guess Workbench internals from the UI. They should use:
* [For agents](/getting-started/for-agents) for product routing and terminology.
* [TextMine MCP server](/integrations/textmine-mcp-server) for direct TextMine tool access.
* [API and developer docs](/integrations/api-and-developer-docs) for Public API V3 reference, examples, MCP setup, and exact operation IDs.
When an external agent needs to modify TextMine state, prefer the API or MCP tool that represents the product action directly. Use Workbench when a person needs to collaborate with the model, inspect intermediate reasoning, upload ad hoc files, or approve a change before it becomes durable.
## Common patterns
* Start with Workbench to explore a messy request, then save the repeatable parts as a workflow.
* Upload files into Workbench for temporary context, then import them into Vault only when they need durable processing and governance.
* Name the source scope in the prompt when it matters: a Vault, folder, selected documents, uploaded files, provider import, or previous table.
* Ask Workbench to use Legislate when the question is about validated, published document data.
* Use records when the answer should become structured business data rather than a one-off response.
# Use files and integrations in Workbench
Source: https://docs.textmine.com/workbench/uploads-and-integrations
Upload files or import from connected MCP providers.
Workbench can use files from several sources.
## Uploaded files
Attach files directly to a Workbench request when you want a quick answer, comparison, template fill, or a later import into Vault. Workbench can answer from uploaded text-like documents without requiring Vault processing.
Uploaded files are treated as temporary Workbench context unless you ask to save them into Vault. When a follow-up request refers to "these files", "the uploaded spreadsheet", or a previous uploaded-source result, Workbench should keep the next action scoped to that uploaded source instead of switching to the whole Vault.
If you ask Workbench to save uploaded files into a Vault, it will use the Vault document pipeline and may ask for approval before mutating Vault content.
## Connected providers
When Workbench MCP providers are enabled, Workbench can list providers, browse folders, list files, and import selected files into the current Workbench chat. Imported provider files become context for the current task; saving them into Vault is a separate approved action.
Workbench MCP providers are different from the hosted TextMine MCP server:
* Workbench MCP providers let a user bring external files into Workbench.
* The TextMine MCP server exposes TextMine tools to external agent clients.
* The ReadMe developer MCP surface helps agents discover API documentation and generated API operations from the developer hub.
Use [Workbench MCP providers](/integrations/workbench-mcp-providers) for provider import behavior, [TextMine MCP server](/integrations/textmine-mcp-server) for external agent tool access, and [API and developer docs](/integrations/api-and-developer-docs) for API setup and examples.
## Good prompts
* "Import the NDA from Google Drive and summarize it."
* "Compare this uploaded spreadsheet with the documents in BU1."
* "Save these uploaded contracts into the Finance Vault."
* "Fill this uploaded Excel template from the values we discussed."
* "Find the latest supplier file in SharePoint, use it as context, and draft a playbook from the obligations."
## Durable versus temporary context
Use uploaded files or provider imports as temporary context when the user is asking a one-off question. Save files into Vault when they need extraction, review, permissions, workflow triggers, audit history, or publication to Legislate.
Agents should make this distinction explicit before taking action. Importing a file into Workbench does not automatically make it a governed Vault document.
# Workflow blocks
Source: https://docs.textmine.com/workflows/blocks
Reference for workflow triggers, conditions, and action blocks.
Workflow blocks are the building units of a TextMine workflow. A workflow starts from a trigger, can route through one or more conditions, and then runs action blocks in order.
Use this page to understand product behavior. For exact JSON schemas, API operation IDs, and examples, use the developer hub's [Public API reference](https://developers.textmine.com/reference) and workflow catalog endpoints.
## How blocks work
Each block has a type, configuration, input context, and output. Later blocks can use earlier outputs through labels or variables.
* **Triggers** decide when the workflow starts.
* **Conditions** decide which route should run.
* **Actions** do work, create outputs, or change TextMine or external systems.
Blocks that create durable effects should be treated carefully by agents. Request confirmation before publishing, sending email, changing assignments, creating records, moving documents, submitting to integrations, or triggering a workflow with business side effects.
## Trigger blocks
| Block | What it does | Use it when |
| -------------------- | --------------------------------------------------------------- | --------------------------------------------------------------------------- |
| `document_imported` | Starts as soon as a document is added to a Vault. | New documents should immediately enter an automation. |
| `document_processed` | Starts when a Vault document finishes processing. | The workflow needs extracted text, tags, or processed document state. |
| `document_published` | Starts when a Vault document reaches the published state. | Publication should trigger follow-up work, reporting, or Legislate handoff. |
| `email_received` | Starts when an email arrives at the workflow's inbound address. | Email should create documents, trigger review, or route attachments. |
| `intake_form` | Creates a public intake form and starts on submission. | External users should submit structured data or files into TextMine. |
| `scheduled` | Runs on a recurring schedule against selected documents. | The process is periodic, such as weekly monitoring or monthly reporting. |
| `manual` | Runs only when explicitly triggered. | A human or API client should decide when the workflow runs. |
Common trigger configuration includes Vault scope, sender access rules, form fields, upload handling, schedule time, timezone, and document selection rules.
## Condition blocks
| Block | What it does | Use it when |
| ------------------------ | -------------------------------------------------------------------- | -------------------------------------------------------------------------- |
| `confidence` | Compares a document confidence score with a threshold. | Low-confidence extraction or review should route differently. |
| `document_type_is` | Checks whether the triggering document has a specific document type. | One workflow handles several document types but needs different routes. |
| `entity_contains` | Checks whether a named entity contains a configured value. | A clause, party, jurisdiction, or extracted field should drive routing. |
| `ai_logic` | Evaluates a natural-language rule and returns a true or false route. | The routing logic needs semantic judgment rather than a fixed field check. |
| `decision_policy_matrix` | Evaluates ordered policy rules and selects a decision route. | The workflow should follow a business policy, matrix, or threshold table. |
Conditions should be as specific as possible. Use deterministic conditions for known fields and AI logic only when the condition needs interpretation.
## Human routing and review actions
| Block | What it does | Notes |
| ------------------ | ------------------------------------------------------ | ---------------------------------------------------------------------------------- |
| `add_reviewer` | Adds users as reviewers on the matched document. | Use for human review queues. |
| `add_assignee` | Adds users as assignees on the matched document. | Use when work ownership should change. |
| `create_alert` | Creates a task, reminder, or alert. | Can include recipients and due dates. |
| `request_approval` | Pauses the workflow until approvers approve or reject. | Use before publishing, sending, external submission, or other high-impact actions. |
## AI, Workbench, and playbook actions
| Block | What it does | Notes |
| -------------------------- | --------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------- |
| `ai_review` | Runs an AI review pass against extracted tags using an instruction. | Best for validating extraction or review quality. |
| `run_general_action` | Runs an analytical instruction and captures a labeled output. | Use for summaries, classifications, or reusable reasoning outputs. |
| `run_workbench_capability` | Runs Workbench from a natural-language instruction using the workflow's Vault, document, and input scope. | Use when the workflow needs Workbench-style planning or analysis as a step. |
| `run_agent` | Triggers a preconfigured agent and waits for the result. | The agent result becomes workflow output. |
| `apply_playbook` | Runs a playbook against documents and can create a Word artifact. | Use when review rules should be reusable and auditable. |
## Values and data pipeline actions
| Block | What it does | Notes |
| ---------------------------- | -------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `set_values` | Seeds static labeled values for downstream steps. | Useful for controlled lists, constants, or manual inputs. |
| `calculate_values` | Applies formulas, variables, and thresholds to workflow inputs. | Use for numeric or rule-based calculations. |
| `organise_values` | Extracts, normalizes, deduplicates, or reformats values. | Use before records, reports, or API calls. |
| `reconcile_data` | Compares inputs or outputs and reports breaks or exceptions. | Use for matching, variance checks, and exception reports. |
| `compare_to_accepted_output` | Compares generated output with an approved answer, table, CSV, JSON, or Excel file. | Use for QA and regression-style checks. |
| `run_python` | Runs a bounded Python data-analysis script over workflow inputs, labels, arrays, variables, and table outputs. | Use for calculations, table cleanup, exception summaries, and intermediate values. REST and network calls are blocked; use `api_request` for external APIs. |
| `enrich_values` | Searches web or named sources and can download files. | Use when external context is needed. |
| `api_request` | Runs a structured REST API request. | cURL or Python snippets can be imported as configuration, but arbitrary code is not executed. Responses can be captured as text, JSON values, or downloadable files. |
| `webhook` | Calls an outbound webhook or HTTP callback. | Uses the same structured request model as API request and can use secret placeholders in headers or bodies. |
| `tableau_export` | Exports Tableau views or workbooks. | Supports CSV, Excel, PDF, PNG, and PowerPoint outputs where the Tableau integration is configured. |
| `power_bi_export` | Exports Power BI reports. | Supports PDF, PowerPoint, and PNG outputs where the Power BI integration is configured. |
`run_python` exposes workflow context such as `workflow_input`, `collection`, `inputs`, `arrays`, `variables`, `tables`, `workflow_tables`, `vault_tables`, `agent_tables`, `workbench_tables`, and `tables_by_label`. Workflow labels are also exposed as `snake_case` variables with `*_list` arrays. Top-level result keys and public variables can be saved as workflow variables for later blocks.
## Records and analytics actions
| Block | What it does | Notes |
| ------------------------- | --------------------------------------------------------------------- | ---------------------------------------------------------------- |
| `lookup_record` | Searches business records by query or name. | Saves matches under an output label. |
| `create_or_update_record` | Creates or updates a business record and can link workflow documents. | Use when workflow output should become structured business data. |
| `get_analytics_metric` | Reads an analytics metric for a scope and period. | Current catalog includes pages processed. |
| `sample_population` | Selects a fixed or percentage sample and emits a sample register. | Use for audit, QA, and population testing. |
## Document, source, and ingest actions
| Block | What it does | Notes |
| ------------------------------ | ------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------- |
| `workbench_uploaded_files` | Uses files attached in Workbench as workflow input. | Good for turning supervised Workbench work into automation. |
| `upload_to_vault` | Uploads inbound attachments into the workflow Vault. | Common after email triggers. |
| `move_document_to_folder` | Moves the current workflow document scope into a target Vault folder. | Use after intake, classification, approval, or exception routing. |
| `extract_text` | Extracts plain text from submitted files. | Common after intake form upload. |
| `resolve_ecfr_sources` | Extracts CFR/eCFR citations, fetches official eCFR text, and can follow capped one-hop cross-references. | Use before regulatory analysis, comparison, or report steps. |
| `resolve_fca_handbook_sources` | Extracts FCA Handbook references, fetches official provision text, and can follow capped one-hop cross-references. | Use when FCA source text should ground downstream analysis. |
| `search_edgar_filings` | Searches SEC EDGAR filing text and saves matching filings. | Use when filings should feed later workflow steps. |
| `fetch_sec_filings` | Imports recent SEC filings for companies in scope. | Supports filing type and folder behavior configuration. |
| `wait_for_document_processing` | Blocks until in-scope documents finish processing. | Use before steps that need extracted text, tags, or document state. |
| `extract_tables_or_line_items` | Extracts tables or line items from documents. | Can target specific table questions or extract detected line items. |
## Report and template actions
| Block | What it does | Notes |
| ---------------------- | -------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- |
| `fill_template` | Fills a spreadsheet, CSV, text, Word, or form template from workflow data. | Can use AI field matching. |
| `send_report_template` | Generates a report from a predefined template. | Use when a standard report exists; workflow document context is passed into the report step. |
| `send_report_custom` | Generates a report from a free-form prompt. | Use for one-off or flexible report outputs grounded in the current workflow scope. |
## Delivery and integration actions
| Block | What it does | Notes |
| ------------------------- | --------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------- |
| `send_to_legislate` | Publishes in-scope imported documents to Legislate. | Use after validation when documents or data should enter the Legislate archive. |
| `import_from_integration` | Pulls records and attachments from an external system into the workflow Vault. | Current catalog supports ServiceNow. |
| `submit_to_integration` | Pushes outputs to an external system. | Supports SharePoint file uploads and ServiceNow submissions for attaching workflow artifacts, updating records, or both where enabled. |
| `send_email` | Sends an email notification and can attach outputs, reports, documents, templates, or intake attachments. | Use sender modes such as original sender or intake submitter for inbound workflows. |
## Agent guidance
Agents should read workflow catalogs before creating or editing workflow definitions. The TextMine MCP server exposes catalog tools for triggers, conditions, and actions, and the Public API exposes the same catalog for exact parameter schemas.
When designing a workflow, agents should return:
* Trigger kind and scope.
* Conditions and routing assumptions.
* Action sequence and expected outputs.
* Side effects that need human confirmation.
* IDs needed for audit, such as workflow ID, run ID, Vault ID, document IDs, record IDs, and report IDs.
# Create a workflow
Source: https://docs.textmine.com/workflows/create-a-workflow
Design a reusable workflow with a trigger and steps.
Create a workflow when a process should run more than once or should start automatically from a trigger.
Before choosing steps, review [Workflow blocks](/workflows/blocks) for the available triggers, conditions, and actions.
## 1. Choose a trigger
Pick the event that starts the workflow: manual, document imported, document processed, document published, email received, intake form, or scheduled.
Choose the narrowest trigger that matches the business process. If the workflow should only run for a certain Vault, document type, team, intake form, or schedule, capture that in the workflow design instead of relying on manual filtering later.
## 2. Add steps
Add action and condition blocks in the order they should run. Prefer specific actions when available, such as Apply Playbook, Lookup Record, Create/Update Record, Fill Template, Request Approval, Send Email, or API Request. Use Run General Action for reusable AI-led analysis that produces a named output for later steps.
Common building blocks include:
* Wait for document processing before reading extracted text or tags.
* Apply a playbook when the workflow needs reusable review rules.
* Fill a template when the output should become a document.
* Lookup or create a record when the output should become structured business data.
* Request approval before publishing, sending, or making high-impact changes.
* Send to Legislate when validated data or documents should be published to the Legislate archive.
## 3. Configure outputs
Use output labels and variables when later steps need values from earlier steps. Record, template, report, email, and API steps can all depend on prior outputs.
Name outputs for the business meaning of the value, not just the step that produced it. For example, use `counterparty_name`, `renewal_notice_date`, or `risk_summary` rather than `step_2_output`.
## 4. Test and monitor
Run the workflow manually first when possible. Use the builder warnings, preview/test tools, and run history to inspect trigger scope, step status, generated files, playbook applications, approval state, errors, and outputs.
When testing with agents, ask them to summarize:
* The trigger and input documents.
* Each completed, skipped, or failed step.
* Any records, reports, alerts, approvals, emails, or Legislate publications created.
* The identifiers a human should use for audit or follow-up.
## Developer and agent setup
External agents should use product docs for workflow semantics and the developer hub for exact operations. Start with [For agents](/getting-started/for-agents), then use [API and developer docs](/integrations/api-and-developer-docs) for Public API V3 reference, examples, and MCP setup.
Before creating or triggering workflows through API or MCP, agents should check for an existing workflow catalog entry and ask for confirmation if the workflow has external side effects.
# Email, intake forms, and approvals
Source: https://docs.textmine.com/workflows/email-intake-and-approvals
Use external inputs and human review in workflows.
Workflows can accept inputs from email, public intake forms, and approval links.
## Email received
Use an Email Received trigger when documents or requests arrive in a workflow-specific mailbox. Configure sender access, allowed senders or domains, and how the email source should be parsed.
## Intake forms
Use an Intake Form trigger for future submitters who need to provide structured fields or files. Do not use intake forms for files that are already attached to the current Workbench request; use Workbench-uploaded files for that case.
## Approval requests
Use Request Approval when a workflow needs a human decision before continuing. Approval steps can branch, wait asynchronously, and show up in the workflow run history.
# Workflow examples
Source: https://docs.textmine.com/workflows/examples
Workflow patterns for financial services, legal, risk advisory, and procurement teams.
Use these examples as starting points for repeatable TextMine automations. They are deliberately written as patterns rather than rigid templates, because most workflows need to be scoped to a Vault, document type, approval policy, and integration setup.
For the full block catalog, see [Workflow blocks](/workflows/blocks).
## Financial services
Financial services workflows usually prioritize intake control, source evidence, exception handling, and audit-ready outputs.
| Workflow | Trigger | Core blocks | Output |
| --------------------------------- | ------------------------------- | -------------------------------------------------------------------------------------------------- | ---------------------------------------------------------- |
| Credit agreement covenant review | Document imported or manual run | Wait for processing, extract tables, run playbook, calculate values, request approval, send report | Covenant exceptions, source-linked review, approval record |
| SEC filing monitor | Scheduled | Fetch SEC filings, search filings, extract text, run general action, create alert, send email | Filing summary and exceptions for monitored issuers |
| Fund document risk scan | Document processed | Default tags, aggregation tags, entity checks, AI review, report template | Risk summary by fund, party, obligation, and date |
| KYC or onboarding evidence review | Intake form | Upload to Vault, wait for processing, extract entities, reconcile data, create or update record | Reviewed customer/vendor record with evidence links |
| Regulatory change impact check | Scheduled or manual | Enrich values, search filings or sources, run agent, create alerts | Impact summary and affected document list |
### Example: covenant exception workflow
1. Start when a credit agreement or amendment is imported into a lending Vault.
2. Wait for processing and table extraction.
3. Run default tags for borrower, facility amount, dates, governing law, and key covenants.
4. Apply a covenant playbook.
5. Use `calculate_values` to compare extracted thresholds against policy limits.
6. Use `reconcile_data` to flag breaks between agreement text, tables, and accepted policy outputs.
7. Request approval from legal or credit risk for exceptions.
8. Send a report with source passages, table references, and reviewer decisions.
Use this pattern when the answer must be defensible to risk, compliance, or audit teams.
## Legal teams
Legal workflows usually focus on review consistency, negotiation guidance, clause risk, publication, and reusable work product.
| Workflow | Trigger | Core blocks | Output |
| ----------------------------- | ---------------------------- | ----------------------------------------------------------------------------------------- | ------------------------------------------ |
| NDA review and escalation | Document imported | Wait for processing, apply playbook, condition on red flags, request approval, send email | Approved NDA review or escalation packet |
| Contract playbook application | Manual or document processed | Apply playbook, AI review, fill template, send report | Clause-by-clause review memo |
| MSA deviation tracker | Document processed | Default tags, chained tags, compare to accepted output, create record | Deviation register with evidence |
| Signature and execution check | Document imported | Embedded image detection, image tags, entity tags, AI review | Signature/completion checklist |
| Publish validated contracts | Approval completed | Send to Legislate, create alert, update record | Published archive document and audit trail |
### Example: legal playbook review
1. Start manually from Vault or after document processing.
2. Confirm the document type and processing status.
3. Apply the correct playbook for the agreement type.
4. Extract parties, dates, governing law, assignment, termination, liability, confidentiality, and data protection fields.
5. Use chained tags to connect extracted parties to obligations and rights.
6. Route red flags to a reviewer.
7. Generate a report or Word artifact for the matter file.
8. Publish to Legislate after validation if the document should become part of the archive.
Use this pattern when a legal team wants consistent review outputs without turning every review into a bespoke Workbench conversation.
## Risk advisory
Risk advisory workflows usually combine document review, sampling, evidence, and exception reporting.
| Workflow | Trigger | Core blocks | Output |
| ---------------------------- | ------------------------------ | ------------------------------------------------------------------------------- | ------------------------------------------------- |
| Control evidence sample | Scheduled or manual | Sample population, wait for processing, extract tags, AI review, report | Sample register and exceptions |
| Third-party risk review | Intake form or provider import | Upload to Vault, extract entities, enrich values, apply playbook, create record | Risk profile with evidence and recommended action |
| Policy compliance sweep | Scheduled | Select documents, run aggregation tags, decision matrix, create alerts | Compliance exception queue |
| Internal audit document test | Manual | Extract tables, compare to accepted output, reconcile data, send report | Test result with supporting documents |
| Regulatory impact assessment | Scheduled | Enrich values, run agent, lookup records, create alerts | Impacted controls, vendors, contracts, or teams |
### Example: third-party risk workflow
1. Start from an intake form or provider import.
2. Upload questionnaires, contracts, security reports, or policies to Vault.
3. Wait for processing and run default tags.
4. Extract entities such as supplier, parent company, locations, systems, and services.
5. Enrich supplier context where configured.
6. Apply a risk playbook or decision policy matrix.
7. Create or update a supplier risk record.
8. Request approval for high-risk findings.
9. Send a final report and retain source evidence.
Use this pattern when the workflow must produce a structured risk outcome rather than a one-off summary.
## Procurement
Procurement workflows usually prioritize supplier intake, obligation extraction, renewal visibility, approval routing, and integration with source systems.
| Workflow | Trigger | Core blocks | Output |
| ------------------------------ | ----------------------------- | -------------------------------------------------------------------------------- | ------------------------------------------------- |
| Supplier contract intake | Intake form or email received | Upload to Vault, wait for processing, default tags, entity tags, create record | Supplier/contract record with extracted terms |
| Renewal and notice monitoring | Scheduled | Lookup records, check dates, create alert, send email | Renewal queue with owners and dates |
| ServiceNow contract import | Scheduled or manual | Import from integration, wait for processing, extract tags, create/update record | Updated contract inventory from ServiceNow |
| Spend and fee table extraction | Document processed | Extract tables, chained tags, calculate values, report | Pricing and fee summary |
| Procurement approval packet | Manual | Apply playbook, request approval, fill template, send email | Approved procurement memo and supporting evidence |
### Example: renewal monitoring workflow
1. Run on a schedule against a procurement Vault or contract record schema.
2. Look up contracts with renewal, expiry, or notice dates within the review window.
3. Confirm dates against validated Vault tags or record evidence.
4. Create alerts for owners.
5. Send a report to procurement and legal stakeholders.
6. Update records when decisions are made.
Use this pattern when procurement needs durable operational follow-up rather than another spreadsheet reminder.
## Design guidance
Before building any workflow, define:
* The business event that should trigger it.
* The Vault, provider, document type, or record scope.
* The human decisions that require approval.
* The extracted tags, tables, entities, or records that downstream steps depend on.
* The expected output: alert, report, record, email, Legislate publication, external submission, or all of these.
* The run history details needed for audit and debugging.
Agents should present workflows as a proposed sequence first when the workflow will send messages, create records, change assignments, call integrations, or publish documents.
# Workflows overview
Source: https://docs.textmine.com/workflows/overview
Learn how TextMine workflows automate document and data processes with triggers, conditions, approvals, integrations, reports, Python analysis, and run history.
Workflows automate repeatable work in TextMine. A workflow has a trigger, ordered steps, optional conditions, and a run history.
For a block-by-block reference, see [Workflow blocks](/workflows/blocks). For industry-oriented patterns, see [Workflow examples](/workflows/examples).
![TextMine workflow builder showing automation steps for document and data processes]()
## Triggers
Workflows can start manually, on a schedule, when documents are imported, processed, or published, when an email is received, or when someone submits an intake form.
## Actions
Workflow actions can run agents, run Workbench instructions, upload to Vault, move documents between folders, use Workbench-uploaded files, extract text, search EDGAR or fetch SEC filings, resolve eCFR or FCA Handbook sources, wait for document processing, apply playbooks, set or calculate values, run bounded Python analysis, reconcile data, compare against accepted outputs, enrich values, call APIs or webhooks, export Tableau or Power BI outputs, lookup or create records, collect analytics, sample populations, extract tables or line items, run AI reviews, fill templates, send reports, add reviewers or assignees, create alerts, request approvals, send to Legislate, import or submit to integrations, and send email.
## Data analysis and source resolution
Use data pipeline blocks when a workflow needs to transform or test values before later steps run. `run_python` supports bounded Python analysis over workflow inputs, labels, variables, arrays, and table outputs. Use it for calculations, table cleanup, exception summaries, and intermediate values that later reports, records, emails, or approvals can reuse.
Use source-resolution blocks when a workflow needs authoritative source text before analysis. eCFR and FCA Handbook blocks can fetch official provisions from citations and pass that source text into downstream review, report, or comparison steps.
For external APIs, use `api_request` or `webhook` rather than Python. API and webhook blocks use structured request configuration and can capture text, JSON values, or downloadable files.
## Conditions
Conditions include AI logic, decision policy matrices, entity checks, document type checks, and confidence checks.
## Workflows and agents
Agents should treat workflows as durable automation, not chat output. A workflow can create business effects such as sending emails, requesting approvals, publishing data, assigning reviewers, calling APIs, moving documents, creating records, or sending documents to Legislate.
For product guidance, start with this section and [Create a workflow](/workflows/create-a-workflow). For exact API operations, schemas, and examples, use the developer hub's [Public API reference](https://developers.textmine.com/reference) and [agent guide](https://developers.textmine.com/docs/for-agents).
Good agent behavior:
* Read the workflow catalog before creating duplicates.
* Prefer updating or reusing an existing workflow when the process already exists.
* Ask for confirmation before triggering a workflow that sends, publishes, assigns, approves, moves documents, or mutates records.
* Include workflow IDs, run IDs, and relevant document IDs in summaries.
* Use workflow run history when debugging or explaining an automation.
## When to create a workflow
Create a workflow when a process should happen more than once, should run from a trigger, or needs auditable run history. Keep one-off exploration in Workbench until the steps are stable enough to automate.
# Records and integrations in workflows
Source: https://docs.textmine.com/workflows/records-and-integrations
Connect workflows to records, APIs, SEC data, and external systems.
Workflows can update TextMine data and connect to external systems.
## Records
Use Lookup Record to read existing business records during a workflow. Use Create/Update Record to create or update records based on document evidence, extracted values, intake form values, or earlier workflow outputs.
## APIs and webhooks
Use API Request for generic REST calls. Use Webhook for outbound callbacks. Store secrets in the workflow configuration rather than hard-coding credentials in request bodies or headers.
## SEC and EDGAR
Workflow steps can search EDGAR filings or fetch SEC filings for supported filing types. Use these steps when a document process depends on public company filings or recurring filing imports.
## Integrations
Use integration import or submit steps when the workflow should move data or files between TextMine and a connected external system.