Mcp Documentation Server
by andrea9293communityv1.15.1
336
41
Local-first document management and semantic search server for AI coding agents. It stores documents locally, supports uploads and hybrid search, and includes a built-in web dashboard plus REST API on port 3080.
From the registry: Local-first document management and semantic search for AI coding agents
Install & configure
Prerequisites
- requiredNode.js and npmInstall Node.js and npm to run the server via npx.
- optionalOptional Google Gemini API keySet a Gemini API key only if you want to use the AI-powered search tool `search_documents_with_ai`. The server works offline without it.https://aistudio.google.com/app/apikey
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `mcp-documentation-server` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Node.js and npm** — Install Node.js and npm to run the server via npx.
Optional prerequisites:
- Optional Google Gemini API key — Set a Gemini API key only if you want to use the AI-powered search tool `search_documents_with_ai`. The server works offline without it. (https://aistudio.google.com/app/apikey)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","@andrea9293/mcp-documentation-server"]
- optional environment variables:
- `MCP_BASE_DIR`: Base directory for data storage. (example: `<your-base-directory>`)
- `GEMINI_API_KEY`: Google Gemini API key; enables the AI-powered search tool `search_documents_with_ai`. (example: `<your-gemini-api-key>`)
- `MCP_EMBEDDING_MODEL`: Embedding model name for local embeddings. (example: `<your-embedding-model>`)
- `MCP_CACHE_ENABLED`: Enable or disable the LRU embedding cache. (example: `<true-or-false>`)
- `START_WEB_UI`: Set to false to disable the built-in web UI. (example: `<true-or-false>`)
- `WEB_HOST`: Bind address for the web UI. (example: `<your-web-host>`)
- `WEB_PORT`: Port for the web UI. (example: `<your-web-port>`)
- `MCP_STREAMING_ENABLED`: Enable streaming reads for large files. (example: `<true-or-false>`)
- `MCP_STREAM_CHUNK_SIZE`: Streaming buffer size in bytes. (example: `<your-stream-chunk-size-bytes>`)
- `MCP_STREAM_FILE_SIZE_LIMIT`: File size threshold in bytes at which streaming is used. (example: `<your-file-size-limit-bytes>`)
Note: The MCP server uses stdio. It also starts a built-in web UI and REST API by default on the configured host/port (default 127.0.0.1:3080), but the MCP client configuration shown in the README is the stdio npx launch command.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: The MCP server uses stdio. It also starts a built-in web UI and REST API by default on the configured host/port (default 127.0.0.1:3080), but the MCP client configuration shown in the README is the stdio npx launch command.
Tools (12)
add_document
Add a document with title, content, and optional metadata
list_documents
List all documents with metadata and content preview
get_document
Retrieve the full content of a document by ID
delete_document
Remove a document, its chunks, database entries, and associated files
process_uploads
Process all files in the uploads folder with chunking and embeddings
get_uploads_path
Return the absolute path to the uploads folder
list_uploads_files
List files in the uploads folder with size and format info
get_ui_url
Return the Web UI URL
search_documents
Semantic vector search within a specific document
search_all_documents
Hybrid full-text and vector search across all documents
get_context_window
Return neighboring chunks around a given chunk index
search_documents_with_ai
AI-powered search using Gemini
Environment variables
GEMINI_API_KEYrequiredGoogle Gemini API key used to enable AI searchMCP_BASE_DIRBase directory for data storageMCP_EMBEDDING_MODELEmbedding model name for local embeddingsSTART_WEB_UIWhether to start the built-in web interfaceWEB_HOSTBind address for the web UIWEB_PORTPort for the web UIPackages
@andrea9293/mcp-documentation-server
npmv1.15.1transport: stdio
Details
- Language
- TypeScript
- License
- MIT
- Released
- Jun 23, 2026
- Last commit
- Jun 23, 2026
- Registry name
- io.github.andrea9293/mcp-documentation-server