TextMine Developer Hub

Build with TextMine Public API v3, connect agent clients through MCP, and provision autonomous third-party agent workspaces with service-account API keys.

Start here

Goal	Best next page
Make the first API calls	Public API Quickstart
Understand the product model	Core Concepts
Choose between ReadMe MCP and TextMine MCP	ReadMe and TextMine MCP Quickstart
Give an agent canonical instructions	For Agents
Provision a third-party agent organization	Agent self-service signup
Load fuller agent context	Agent Context Pack
Configure Codex	Codex MCP Setup
Configure Claude	Claude MCP Setup
Configure Cursor or Windsurf	Cursor and Windsurf MCP Setup
Configure ChatGPT/OpenAI agents	ChatGPT and OpenAI MCP Setup
Browse exact endpoints and schemas	API Reference
Read agent-friendly project context	llms.txt
Learn the full product docs	docs.textmine.com
Visit the main TextMine site	textmine.com

Public API v3

Public API v3 is the default version for new integrations. It covers the current TextMine surfaces: Vault documents, search, document types, tags, metadata, assignees, users, integrations, workflows, agents, reports, alerts, exports, business records, and API keys.

Current UK multitenant base URL:

https://public-api.textmine.com

Endpoint paths start at /v3/.... Use the environment-specific Public API base URL returned by discovery or provided by TextMine for dedicated deployments.

Authenticate live requests with a TextMine API key:

Authorization: Bearer tm_your_api_key

If a platform owns the Authorization header, use:

X-TextMine-Api-Key: tm_your_api_key

First calls:

curl https://public-api.textmine.com/v3/health \
  -H "Authorization: Bearer tm_your_api_key"

curl https://public-api.textmine.com/v3/vaults \
  -H "Authorization: Bearer tm_your_api_key"

Existing customers can create or request keys through their organization access flow. Autonomous third-party agents in the multitenant, non-enterprise deployment should use Agent self-service signup to create a dedicated organization and encrypted service-account key.

Common integration paths

Import documents into Vault

Use direct upload when your integration already has file bytes. Use provider import when files live in Google Drive, SharePoint, Notion, or another connected source. Start with Document Ingestion, then use the API Reference for exact request bodies.

Work with validated data

Use Vault for active processing, extraction, search, metadata, tags, reports, and workflow automation. Use Legislate Archive when the task is about validated documents that have been published from Vault into Legislate.

Automate repeatable work

Use Reports, Workflows, and Agents when a process should run repeatedly or produce structured output. Workflows can run from schedules, document events, email, intake forms, or manual triggers.

Build with users and access in mind

Use Vaults and Access and Users, Teams, and Assignees before assigning reviews, approvals, or workflow steps. Agents should discover IDs from the API instead of guessing or hard-coding them.

Product map

Surface	Use it for
Vault	Upload, import, process, search, classify, tag, review, report on, and automate documents.
Legislate	Review the validated team-document archive after documents are published from Vault.
Workbench	Explore questions, upload or import files, and orchestrate multi-step work inside the app.
Workflows	Turn repeatable document and data processes into triggers, actions, conditions, and run history.
Playbooks	Apply reusable review guidance, policies, checklists, and negotiation logic to documents.
Records	Convert document evidence into structured business data and relationships.
Agents	Run configurable AI work directly, from Workbench, or inside workflows.
Scribe	Generate documents and templates from fields and conditional logic.
Integrations	Import provider files, call APIs/webhooks, submit to external systems, and expose TextMine through MCP.

Agent and MCP guidance

TextMine has two MCP surfaces:

Surface	URL	Use
ReadMe MCP	https://developers.textmine.com/mcp	Developer docs, OpenAPI schemas, and generated API-reference tools.
TextMine MCP	https://app.textmine.com/mcp	Hosted TextMine product tools backed by Public API v3.

Recommended agent order:

Discover available Vaults with listVaults.
Discover document types, tags, metadata, users, integrations, workflow actions, and workflow triggers before mutating data.
Prefer provider browsing endpoints before provider import.
Prefer stable operation IDs from the API Reference over guessed URLs.
Use the caller's own bearer key so TextMine permissions, Vaults, and integrations resolve correctly.

The v3 reference is generated from OpenAPI and is intended to be readable by humans and agents. For broader product context, use docs.textmine.com; for endpoint-level work, use this developer hub.