Cloudflare AutoRAG MCP Server ๐Ÿ“ก

โš ๏ธ Deprecated

This server is deprecated. AutoRAG has been superseded by Cloudflare AI Search, and the unified Cloudflare MCP server at mcp.cloudflare.com/mcp already covers AI Search (along with the rest of the Cloudflare API).

All new work should move to the unified server:

{
	"mcpServers": {
		"cloudflare-api": {
			"url": "https://mcp.cloudflare.com/mcp"
		}
	}
}

That server uses Code Mode โ€” two generic tools (search and execute) that give agents access to the full Cloudflare API through code execution. It supports both OAuth (connect to the URL and authorize) and Cloudflare API tokens (send as a bearer token). See cloudflare/mcp for details.

What about my existing AutoRAG instances?

  • Existing AutoRAG instances have been migrated to AI Search. They remain reachable via both the legacy AutoRAG endpoints and the new AI Search endpoints โ€” the underlying instance is the same, just exposed through both APIs.
  • Because your instances are already available through AI Search, they are accessible through the unified mcp.cloudflare.com/mcp server (which covers AI Search) without any data migration on your part.
  • The AutoRAG REST API endpoints are no longer recommended for new work. Refer to Migrate from AutoRAG REST API for moving client code to AI Search.
  • This MCP server continues to respond for now, but will be retired.

The tools below still function. Please migrate at your earliest convenience.

This is a Model Context Protocol (MCP) server that supports remote MCPconnections, with Cloudflare OAuth built-in.

It integrates tools powered by the Cloudflare AutoRAG API to allow you to access and query your account's AutoRAG instances.

๐Ÿ”จ Available Tools

Currently available tools:

Tool Description
list_rags Lists AutoRAGs with pagination support
search Searches documents in a specified AutoRAG using a query (URL, title, or snippet)
ai_search Performs AI-powered search on documents in a specified AutoRAG

Prompt Examples

  • List all AutoRAGs in my account.
  • Search for documents in AutoRAG with ID 'rag123' using the query 'cloudflare security'.
  • Perform an AI search in AutoRAG with ID 'rag456' for 'best practices for vector stores'.

Access the remote MCP server from any MCP Client

The following setup documentation is retained for existing users. New users should follow the migration path to mcp.cloudflare.com/mcp documented above.

If your MCP client has first class support for remote MCP servers, the client will provide a way to accept the server URL (https://autorag.mcp.cloudflare.com) directly within its interface (for example in Cloudflare AI Playground).

If your client does not yet support remote MCP servers, you will need to set up its respective configuration file using mcp-remote (https://www.npmjs.com/package/mcp-remote) to specify which servers your client can access.

Replace the content with the following configuration:

{
	"mcpServers": {
		"cloudflare": {
			"command": "npx",
			"args": ["mcp-remote", "https://autorag.mcp.cloudflare.com/mcp"]
		}
	}
}

Once you've set up your configuration file, restart MCP client and a browser window will open showing your OAuth login page. Proceed through the authentication flow to grant the client access to your MCP server. After you grant access, the tools will become available for you to use.

Interested in contributing, and running this server locally? See CONTRIBUTING.md to get started.

MCP Server ยท Populars

MCP Server ยท New

    LeulAria

    Aria Icons

    MCP server for 340k SVG icons

    Community LeulAria
    knowall-ai

    Reverie โ€” graph memory that dreams

    Memory management MCP server for AI agents using Neo4j knowledge graphs

    Community knowall-ai
    keploy

    Key Highlights

    Open-source platform for creating safe, isolated production sandboxes for API, integration, and E2E testing.

    Community keploy
    hermes-labs-ai

    Fidelis Memory

    Zero-LLM agent memory for Claude Code and AI agents: local-first BM25, dense-vector, and reciprocal-rank-fusion retrieval. Returns original passages verbatim by default. Available on PyPI as fidelis-memory. MIT.

    Community hermes-labs-ai
    n24q02m

    Better Code Review Graph

    Knowledge graph for token-efficient code reviews -- semantic search and call-graph resolution across your codebase.

    Community n24q02m