MCP (Model Context Protocol)
The Model Context Protocol (MCP) is an open standard that lets AI assistants such as Claude connect to external tools and data sources, including documentation, through a common server interface.
For docs, MCP means an AI coding assistant can query your documentation directly. A developer using Claude Code or Cursor connects to a docs MCP server, and the assistant searches the docs and answers with current, product-specific information instead of stale training data.
An MCP server exposes capabilities as tools. A documentation MCP server typically exposes a search tool over the indexed docs, so the assistant retrieves relevant passages on demand. The mechanics resemble RAG, but the request comes from the user's own AI client.
Biel.ai provides a hosted MCP server for docs sites, so teams can offer this without running infrastructure: users add the server URL to their AI client and get grounded answers from the docs inside their editor.
Frequently asked questions
Who created MCP?
Anthropic released MCP as an open standard in November 2024. It has since been adopted broadly across AI clients and tooling from other vendors.
What is a docs MCP server?
An MCP server that exposes documentation search as a tool. AI assistants connected to it can look up current docs while answering, instead of relying on training data.
Do users need to install anything?
They add the server URL to an MCP-capable client such as Claude Code, Claude Desktop, or Cursor. A hosted server requires no local installation.