Connect your AI agent

CaptableAI runs a hosted Model Context Protocol (MCP) server. Connect Claude, Cursor, ChatGPT, or VS Code and manage your cap table in plain language — upload documents, run extraction, edit objects, model scenarios, and commit. Learn about MCP →

Server endpoint

Point any MCP client at this endpoint. Authentication is OAuth — your client opens a browser to sign in the first time, then reuses the session.

https://mcp.captable.ai/mcp

Set up your client

{
  "mcpServers": {
    "captable-ai": {
      "url": "https://mcp.captable.ai/mcp"
    }
  }
}

https://mcp.captable.ai/mcp

{
  "servers": {
    "captable-ai": {
      "type": "http",
      "url": "https://mcp.captable.ai/mcp"
    }
  }
}

https://mcp.captable.ai/mcp

{
  "mcpServers": {
    "captable-ai": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.captable.ai/mcp"]
    }
  }
}

What your agent can do

• Read your cap table — Capitalization, stakeholders, convertibles, and metrics on demand.
• Upload & extract — Drop in legal documents and run OCF extraction.
• Edit & commit — Review extracted objects, fix them, and commit to the cap table.
• Model scenarios — Run pro-forma financing, conversion, warrant, and option-pool scenarios.
• Export — Generate an Excel cap table or a shareholder register PDF.

Uploading documents — the optimal flow

CaptableAI builds your cap table by reading your legal documents. The pipeline is automated, but a few habits make extraction faster and more accurate — share these with your agent.

1. 1 Upload everything first — Add all of your cap table documents up front, in one batch where possible — incorporation docs, SAFEs, stock and option agreements, board consents. Supported formats: PDF, DOCX, XLSX, DOC, and images. Files up to 25 MB go through the MCP; use the CLI for larger files (up to 50 MB).
2. 2 Wait for classification — Each document is OCR’d and then classified — the system detects its type and its effective date. Do not trigger extraction until every document has reached the CLASSIFIED status.
3. 3 Check the classification dates — The effective date set during classification determines the order documents are processed in. If a date looks wrong, correct it before triggering extraction so your cap table is built in the right chronological order.
4. 4 Trigger extraction once — Extraction does not start automatically for agent or CLI uploads — trigger it a single time after all documents are classified. Everything queued is picked up in that run.
5. 5 Let the FIFO queue run — Documents are extracted one at a time, in ascending order of their classification date (earliest first), so later events build on earlier ones. Expect a few minutes per document — don’t re-trigger while it works.
6. 6 Validation runs automatically — Once the last document finishes extraction, a company-wide validator runs across all extracted objects to repair cross-document issues. It only kicks off after every document has reached validation, so there’s nothing to trigger.
7. 7 Review, then commit — When all documents reach PROCESSED, review the draft objects and commit. The first commit activates your cap table.

Prefer the terminal? @fairmint/captable-ai CLI runs the same commands for scripts and CI. Files larger than 25 MB must be uploaded with the CLI.

About CaptableAI · Contact