Inkeep MCP Server
Search your product docs and content via Inkeep's RAG, exposed as an MCP tool.
Add to your client
Copy the config for your MCP client and paste it into its config file.
git clone https://github.com/inkeep/mcp-server-python.gitPaste into ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"inkeep-mcp-server": {
"command": "uv",
"args": [
"--directory",
"<YOUR_INKEEP_MCP_SERVER_ABSOLUTE_PATH>",
"run",
"-m",
"inkeep_mcp_server"
],
"env": {
"INKEEP_API_BASE_URL": "https://api.inkeep.com/v1",
"INKEEP_API_KEY": "<YOUR_INKEEP_API_KEY>",
"INKEEP_API_MODEL": "inkeep-rag",
"INKEEP_MCP_TOOL_NAME": "search-product-content",
"INKEEP_MCP_TOOL_DESCRIPTION": "Retrieves product documentation about Inkeep. The query should be framed as a conversational question about Inkeep."
}
}
}
}Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf
Before you start
- An Inkeep account to manage and provide the RAG (https://inkeep.com)
- The uv Python project manager (https://github.com/astral-sh/uv)
- An Inkeep API key created from the Inkeep portal (Projects > Integrations > Create Integration > API)
About Inkeep MCP Server
Inkeep's official MCP server lets an MCP client retrieve answers from your product documentation and content through Inkeep's RAG service. After cloning the repo, setting up a uv virtual environment, and creating an Inkeep API key, you register the server in your MCP client's config (e.g. claude_desktop_config.json). The server exposes a single retrieval tool whose name and description are set via environment variables, so it surfaces as a domain-specific search tool (default: search-product-content) the model can call with conversational questions.
Tools & capabilities (1)
search-product-contentConfigurable retrieval tool (default name from INKEEP_MCP_TOOL_NAME). Retrieves product documentation/content from Inkeep's RAG. The query should be framed as a conversational question, and the tool name/description are customizable via the INKEEP_MCP_TOOL_NAME and INKEEP_MCP_TOOL_DESCRIPTION environment variables.
What this server can do
Inkeep MCP Server provides tools for these capabilities — tap one to see every MCP server that does the same:
When to use it
- Let Claude Desktop answer questions grounded in your own product documentation
- Provide an MCP-accessible search over a company's docs and product content
- Build doc-aware assistant workflows backed by Inkeep's RAG
Security notes
Requires an Inkeep API key (created from the Inkeep portal under Integrations > API). The key is passed via the INKEEP_API_KEY environment variable in your MCP client config; keep it secret and avoid committing it.
Inkeep MCP Server FAQ
Do I need an Inkeep account?
Yes. You need an Inkeep account to manage and provide the RAG, and you must create an API key from the Inkeep portal (Projects > Integrations > Create Integration > API).
How is the search tool named?
The tool name and description are configurable through the INKEEP_MCP_TOOL_NAME and INKEEP_MCP_TOOL_DESCRIPTION environment variables. The README's example uses search-product-content.
What runtime does it use?
Python, run via the uv project manager. You clone the repo, create a uv venv, install dependencies from pyproject.toml, and launch it with `uv run -m inkeep_mcp_server`.
Alternatives to Inkeep MCP Server
Compare all alternatives →
Official MCP reference server that fetches a URL and returns its content as clean Markdown, with chunking.
Official Firecrawl MCP server — scrape, crawl, map, search, and structured extraction for any LLM client.
Exa's neural web search and crawling MCP server — runs locally via npx or as a hosted remote endpoint.
Compare Inkeep MCP Server with: