mcp-documentation-server

Overview

The MCP Documentation Server is a sophisticated tool for managing documents locally with advanced search capabilities. It uses an embedded Orama vector database for hybrid full-text and vector similarity search, with intelligent parent-child chunking to preserve context. The server includes a built-in web UI for document management and offers optional AI-powered search through Google Gemini integration. All data is stored locally in ~/.mcp-documentation-server/ with no external dependencies required.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• add_document

  Add a document (title, content, optional metadata)

  Try:Add a new document about React best practicesTry:Create a document titled 'Project Guidelines' with team policies
• list_documents

  List all documents with metadata and content preview

  Try:Show me all documents in the knowledge baseTry:List all documents with their titles and creation dates
• get_document

  Retrieve the full content of a document by ID

  Try:Get the full content of document 123Try:Retrieve the document with ID abc-123
• delete_document

  Remove a document, its chunks, database entries, and associated files

  Try:Delete document with ID 456Try:Remove the outdated marketing strategy document
• process_uploads

  Process all files in the uploads folder (chunking + embeddings)

  Try:Process all uploaded filesTry:Handle the new documents in the uploads folder
• get_uploads_path

  Returns the absolute path to the uploads folder

  Try:Where are the uploaded files stored?Try:Show me the path to the uploads directory
• list_uploads_files

  Lists files in the uploads folder with size and format info

  Try:What files are in the uploads folder?Try:Show me all pending uploads with their sizes
• get_ui_url

  Returns the Web UI URL (e.g. http://localhost:3080)

  Try:What's the URL for the documentation dashboard?Try:Open the web interface
• search_documents

  Semantic vector search within a specific document

  Try:Search for 'authentication methods' in document 789Try:Find all references to API keys in the technical doc
• search_all_documents

  Hybrid (full-text + vector) cross-document search

  Try:Search across all documents for 'performance optimization'Try:Find information about database migrations in the knowledge base
• get_context_window

  Returns a window of chunks around a given chunk index

  Try:Show me context around chunk 123Try:Get neighboring chunks for position 456
• search_documents_with_ai

  AI-powered search using Gemini (requires GEMINI_API_KEY)

  Try:Use AI to analyze all documents about user feedbackTry:Find insights about recent feature requests using AI

Comparable tools

Installation

{
  "mcpServers": {
    "documentation": {
      "command": "npx",
      "args": [
        "-y",
        "@andrea9293/mcp-documentation-server"
      ],
      "env": {
        "MCP_BASE_DIR": "/path/to/workspace",
        "GEMINI_API_KEY": "your-api-key-here",
        "MCP_EMBEDDING_MODEL": "Xenova/all-MiniLM-L6-v2",
        "START_WEB_UI": "true",
        "WEB_PORT": "3080"
      }
    }
  }
}

npm install -g @andrea9293/mcp-documentation-server

FAQ

What file formats does the server support?

The server supports .txt, .md, and .pdf files for upload and processing.

Is the web UI optional?

Yes, you can disable the web UI by setting START_WEB_UI=false in your environment configuration.

Compare mcp-documentation-server with