
by Matra
Matra Docs MCP serves the Matra editor framework documentation to AI assistants over the Model Context Protocol. It lets Claude, Cursor, Codex and other clients answer Matra questions from the docs instead of from memory.
PluginContact for pricing
Step 1: In Claude Code, run claude mcp add matra -- npx -y @matrajs/mcp.
Step 2: For Claude Desktop, add a matra entry under mcpServers in claude_desktop_config.json with command npx and args -y @matrajs/mcp. Cursor uses the same object in .cursor/mcp.json.
Step 3: For Codex, add an mcp_servers.matra block with command npx to ~/.codex/config.toml.
Step 4: For clients that speak HTTP, run npx -y @matrajs/mcp --http 3333 and point the client at http://localhost:3333/mcp.
Step 5: Ask the assistant a Matra question. It calls search_docs, reads the page it needs with read_doc and answers from it.
No. Every documentation page ships inside the package, and nothing is fetched while it runs.
No. It declares three tools, all marked read-only and idempotent, and no prompts.
Claude Code, Claude Desktop, Cursor and Codex are documented, as is any client that speaks HTTP.
The core is MIT and free. Pro and Business are paid plans for AI, collaboration and version history, priced flat rather than per seat.
Requirements
MediaWiki MCPMediaWiki MCP Server lets LLM clients read and edit any MediaWiki wiki through MCP tools. It is made by Professional Wiki.
Neruva Agent MCPNeruva is a memory layer for coding agents, delivered as an MCP server backed by a hosted API. Agents recall earlier decisions, log what they do and leave handoff notes between sessions.DocVet MCPdocvet is a Python command-line tool that checks docstrings for missing sections, staleness and rendering problems, and it can run as an MCP server for AI agents.