jdocmunch-mcp

Overview

jDocMunch is a sophisticated MCP server that transforms how AI agents interact with documentation by shifting from file-based to section-based access. It indexes documentation once using heading hierarchy and section structure, then provides precise access to only the content an agent needs. This approach dramatically reduces token consumption, eliminates irrelevant context, and improves agent reliability. The server implements the jMRI-Full specification, supporting discovery, search, retrieval, and metadata operations with batch capabilities, hash-based drift detection, and byte-offset addressing.

Try asking AI

After installing, here are 7 things you can ask your AI assistant:

• index_local{ "path": "/path/to/docs" }

  Index a local folder of documentation

  Try:Index my project documentation at /docs/project
• index_repo{ "url": "owner/repo" }

  Index documentation from a GitHub repository

  Try:Index the documentation from repo owner/example
• get_toc{ "repo": "owner/repo" }

  Get the table of contents for a documentation repository

  Try:Show me the table of contents for the docs
• get_toc_tree{ "repo": "owner/repo" }

  Get the hierarchical table of contents tree

  Try:Show me the full documentation structure as a tree
• get_document_outline{ "repo": "owner/repo", "doc_path": "docs/config.md" }

  Get the outline structure of a specific document

  Try:Outline the structure of the configuration documentation
• search_sections{ "repo": "owner/repo", "query": "authentication" }

  Search for documentation sections matching a query

  Try:Search for sections about authentication in the docs
• get_section{ "repo": "owner/repo", "section_id": "owner/repo::docs/config.md::authentication#1" }

  Retrieve the content of a specific documentation section

  Try:Get the authentication section from the config documentation

Comparable tools

Installation

pip install jdocmunch-mcp

{
  "mcpServers": {
    "jdocmunch": {
      "command": "uvx",
      "args": ["jdocmunch-mcp"]
    }
  }
}

{
  "mcpServers": {
    "jdocmunch": {
      "command": "uvx",
      "args": ["jdocmunch-mcp"],
      "env": {
        "GITHUB_TOKEN": "ghp_...",
        "ANTHROPIC_API_KEY": "sk-ant-..."
      }
    }
  }
}

FAQ

How does jDocMunch compare to traditional documentation retrieval methods?

jDocMunch reduces token consumption by up to 95% compared to traditional file-based approaches. Instead of loading entire documents, it indexes by section and retrieves only the relevant content, making documentation exploration much more efficient.

What file formats does jDocMunch support?

It supports Markdown/MDX/HTML/RST/AsciiDoc for heading-based parsing, OpenAPI tags for structure-based parsing, and Jupyter notebooks for cell-based parsing.

Is jDocMunch free for commercial use?

jDocMunch is free for non-commercial use only. Commercial use requires a paid license with options ranging from $29 (Builder) to $499 (Platform) depending on the number of developers and deployment scope.

Compare jdocmunch-mcp with