Ragdoll AI MCP Server

What is Ragdoll AI MCP Server?

A Model Context Protocol (MCP) server that enables querying Ragdoll AI knowledge bases directly from LLM client applications like Cursor, Windsurf, and Cline.

How to use Ragdoll AI MCP Server?

Install via npx -y ragdoll-mcp-server or clone the repo and run bun run index.ts. Set environment variables RAGDOLL_API_KEY and RAGDOLL_KNOWLEDGE_BASE_ID. Configure the server in your IDE's MCP settings (e.g., mcp.json for Cursor, mcp_config.json for Windsurf, cline_mcp_settings.json for Cline).

Key features of Ragdoll AI MCP Server

• Query Ragdoll AI knowledge bases via MCP.
• Optional topK and rerank parameters.
• NPX installation for quick setup.
• Supports Cursor, Windsurf, and Cline.
• Uses Bun runtime for execution.
• MIT licensed.

Use cases of Ragdoll AI MCP Server

• Integrate Ragdoll knowledge bases with Cursor IDE.
• Query knowledge base content from Windsurf.
• Use with Cline for programmatic access.
• Retrieve ranked results from a knowledge base.

FAQ from Ragdoll AI MCP Server

What does this server do?

It acts as a bridge between LLM clients and Ragdoll AI knowledge bases, allowing users to query their knowledge bases using natural language through the Model Context Protocol.

What are the runtime requirements?

Bun runtime version 1.2.1 or later is required. The server also needs a Ragdoll AI API key and a knowledge base ID.

How do I authenticate?

Authentication is done via the RAGDOLL_API_KEY environment variable. The key is passed to the server when starting it or through the IDE's MCP configuration.

What parameters can I use in a query?

Required: query (string). Optional: topK (number, 1‑10) and rerank (boolean).

Which IDEs are supported?

Cursor, Windsurf, and Cline are explicitly documented with configuration examples. Other MCP‑compatible clients may also work.