Skip to main content

Connect your AI assistant

TridentStack Control publishes a Model Context Protocol (MCP) server for its documentation. Add it to an MCP-compatible AI assistant, such as Claude or Cursor, and the assistant can search and read the TridentStack Control documentation directly. Its answers stay grounded in our current guides, API reference, changelog, and roadmap.

The server is read-only and public. It exposes no account or fleet data, and it needs no sign-in or API key.

Server URL​

https://docs.tridentstack.com/mcp

The transport is Streamable HTTP.

Add it to your assistant​

Claude Code​

claude mcp add --transport http tridentstack-docs https://docs.tridentstack.com/mcp

Claude Desktop​

  1. Open Settings, then Connectors.
  2. Add a custom connector.
  3. Set the URL to https://docs.tridentstack.com/mcp.
  4. Save, then start a new chat.

Cursor​

Add the server to your mcp.json:

{
  "mcpServers": {
    "tridentstack-docs": {
      "url": "https://docs.tridentstack.com/mcp"
    }
  }
}

Other assistants​

Any client that supports a remote MCP server over Streamable HTTP can use the same URL.

What the assistant can do​

Once connected, the assistant has three tools:

  • search_docs: full-text search across the documentation. Returns ranked results with titles, links, and snippets.
  • get_doc: read a specific documentation section in full.
  • list_docs: browse the documentation catalog.

Ask a normal question, such as "How do deployment rings work in TridentStack Control?" or "Which API endpoint lists agents?" The assistant searches the docs, reads the relevant sections, and answers with links back to the source pages.

Prefer a plain file?​

If your tool does not speak MCP, the same content is published as plain text:

  • llms.txt: a short index of the documentation.
  • llms-full.txt: the entire documentation as a single Markdown file.