Skip to content

Docs for AI (MCP / llms.txt)

You can connect the INCY documentation to AI assistants (Claude Code, Codex, Cursor, etc.) so they automatically pull the current subscription formats, headers, deep links, and Premium API — and write your integration faster and more correctly, with no manual copy-paste.

There are two ways: an MCP server (a tool for the AI) and llms.txt (a machine-readable index).

GitMCP is a free, public MCP server that reads documentation straight from our GitHub repository. Nothing to install — just add the URL.

Server URL:

https://gitmcp.io/INCY-DEV/incy-docs

Claude Code

claude mcp add --transport http incy-docs https://gitmcp.io/INCY-DEV/incy-docs

Cursor / Codex / other clients

Add the server to the client's MCP config:

{
  "mcpServers": {
    "incy-docs": {
      "url": "https://gitmcp.io/INCY-DEV/incy-docs"
    }
  }
}

Once connected, the AI gets tools to search and read the INCY docs. For example, given "build an integration: hide the subscription URL and add a banner", the assistant will find the hide-url and banner-text headers in the docs and apply them correctly.

Option 2. llms.txt

For tools that support the llms.txt standard, we publish two files:

File Contents
llms.txt Index: every documentation section with links and a short description
llms-full.txt The entire documentation in one file (full text)

Just give the assistant the https://docs.incy.cc/llms-full.txt link and it gets the whole documentation at once.

What you can ask the AI

  • "Which subscription HTTP headers does INCY support, and what does each one do?"
  • "Show the deep-link format for importing a subscription"
  • "How do I configure fragmentation and domain fronting via the Premium API?"
  • "Build an example full Xray config with routing for INCY"

The documentation is bilingual (RU by default, EN as suffixed copies), and llms.txt / GitMCP update automatically whenever the documentation changes.