general MCP Server
The leading, most token-efficient MCP server for documentation exploration and retrieval via structured section indexing
Discovered via github-seeds and last synced 3mo ago.
1. Install the package
uvx jdocmunch-mcp
2. Add to claude_desktop_config.json
{
"mcpServers": {
"jdocmunch-mcp": {
"command": "npx",
"args": [
"jdocmunch-mcp"
]
}
}
}Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)
Purpose
Index a local documentation folder
Full content of one section
`.md`, `.markdown`
Section summaries via Claude Haiku
One-command onboarding: detect clients, write config, install policy, hooks, index
Index a GitHub repository’s docs
Batch content retrieval
`.mdx`
Traditional approach
Print or install the Doc Exploration Policy (`--install global\
List indexed documentation sets
Section + ancestor headings + child summaries
`.rst`
OpenAI embeddings (text-embedding-3-small)
`~/Library/Application Support/Claude/claude_desktop_config.json`
PostToolUse hook handler (reads JSON from stdin)
Nested section tree per document
Detect internal links/anchors that no longer resolve
`.html`
`~/.config/claude/claude_desktop_config.json`
PreCompact hook handler (reads JSON from stdin)
Section hierarchy for one document
Which jcodemunch symbols have matching doc sections
Path
PreToolUse hook handler (reads JSON from stdin)
Flat section list in document order
Remove a doc index
`.adoc`
Force provider: `gemini`, `openai`, `openai-compatible`, `sentence-transformers`, `none`
sentence-transformers model (default: `all-MiniLM-L6-v2`)
Set to `0` to disable anonymous community token savings reporting
GitHub API auth
Purpose
Custom cache path
Extensions
Section summaries via Gemini Flash; also Gemini embeddings
Weighted search returning summaries only
`%APPDATA%\Claude\claude_desktop_config.json`
Purpose
Config
Batch size for `openai-compatible` embeddings (default: `32`)
Dedicated optional API key for `openai-compatible` embeddings
Endpoint URL for `openai-compatible` embeddings
Model for `openai-compatible` embeddings