Connect your AI tool to these docs

A hosted, read-only MCP server that lets your coding assistant search and read these docs and the OpenAPI document.

These docs are available to AI tools as an MCP server:

https://docs.megtrust.com/mcp

Transport: Streamable HTTP.
Read-only, no authentication. It serves the same public pages as this site. It cannot call the MegTrust API and never sees your keys.

Tool	What it does
`search`	Search the docs with a query.
`list_pages`	List every page with its URL.
`get_page`	Read one page as Markdown, by its URL (e.g. `/docs/guides/send-a-payout`).
`get_openapi`	Return the full OpenAPI 3.1 document for the Client API.

To let an agent call the API with your own key, use the local API connector instead.

Claude Code

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

Cursor

Add to .cursor/mcp.json in your project (or ~/.cursor/mcp.json for every project):

.cursor/mcp.json

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

VS Code

Add to .vscode/mcp.json in your workspace:

.vscode/mcp.json

{
  "servers": {
    "megtrust-docs": {
      "type": "http",
      "url": "https://docs.megtrust.com/mcp"
    }
  }
}

Other tools

Any MCP client that supports Streamable HTTP can connect to https://docs.megtrust.com/mcp directly — no headers, no OAuth. Most use the same shape as Cursor:

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

A client that only speaks stdio can reach it through a local bridge such as mcp-remote.

Try it

Ask your assistant:

Using the MegTrust docs, how do I retry a payout safely after a timeout?

It should find Send a payout and answer with the idempotency key rules.

Prefer plain files? See llms.txt.