- Home
- MCP servers
- DocuMCP
DocuMCP
- javascript
6
GitHub Stars
javascript
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"yannicktm-docu-mcp": {
"command": "npx",
"args": [
"-y",
"@myjungle/docu-mcp-server"
]
}
}
}DocuMCP provides two complementary MCP servers that let Claude generate, search, and orchestrate documentation for your codebase. The core DocuMCP Server creates documentation with vector embeddings and semantic search, while the DocuMCP Manager coordinates multiple Claude Code sub-agents for parallel processing. Together, they scale documentation workflows from single files to enterprise applications.
How to use
Set up Claude Desktop to run both DocuMCP Server and DocuMCP Manager, then configure them to share a single vector database. You will spawn agents to work on documentation tasks in parallel, monitor their status, and retrieve results. Use the core server for generating documentation, user guides, and diagrams, and use the manager to orchestrate multi-agent runs for large codebases. All results are stored in a shared vector store so agents can collaborate seamlessly.
How to install
Prerequisites: install Node.js 20.11.24+ and have Claude Desktop installed. You may also run external vector databases if you choose.
Install or configure the two MCP servers in Claude Desktop using the following setups.
{
"mcpServers": {
"docu-mcp": {
"command": "npx",
"args": ["-y", "@myjungle/docu-mcp-server"]
}
}
}
Advanced configuration and start
Configure Claude Desktop to load both servers and ensure they share the same vector database settings. You will run the core DocuMCP Server and the DocuMCP Manager with appropriate environment variables to enable the shared vector store and embedding providers.
{
"mcpServers": {
"docu-mcp": {
"command": "node",
"env": {
"VECTOR_DB_PROVIDER": "qdrant",
"QDRANT_URL": "http://localhost:6333",
"EMBEDDING_PROVIDER": "ollama",
"EMBEDDING_MODEL": "bge-m3:latest",
"EMBEDDING_DIMENSION": "1024",
"OLLAMA_URL": "http://localhost:11434"
},
"args": ["/absolute/path/to/DocuMCP/mcp/dist/index.js"]
},
"docu-mcp-manager": {
"command": "node",
"env": {
"VECTOR_DB_PROVIDER": "qdrant",
"QDRANT_URL": "http://localhost:6333",
"EMBEDDING_PROVIDER": "ollama",
"EMBEDDING_MODEL": "bge-m3:latest",
"EMBEDDING_DIMENSION": "1024",
"OLLAMA_URL": "http://localhost:11434",
"SUB_AGENT_MODEL": "claude-3-7-sonnet-latest"
},
"args": ["/absolute/path/to/DocuMCP/manager/dist/index.js"]
}
}
}
Available tools
read_file
Read the contents of a file from the project.
write_file
Write data to a file in the project.
create_directory
Create a new directory in the project.
read_directory
List and read files within a directory.
search_codebase
Search code across the repository to locate relevant snippets or definitions.
search_documentation
Search generated or existing documentation for fast access.
search_diagram
Find and retrieve architectural diagrams.
search_user_guide
Locate user guides and related documentation.
generate_documentation
Generate documentation from code and diagrams.
generate_user_guide
Create user guides from code and diagrams.
explain_code
Provide explanations for code snippets.
generate_diagram
Create architectural diagrams from code structure.
merge_diagram
Merge changes into existing diagrams.
index_file
Index a single file for fast retrieval in the vector store.
index_directory
Index an entire directory for comprehensive search.
merge_documentation
Merge new documentation into existing documentation sets.
spawn_agent
Create Claude Code sub-agents to handle documentation tasks in parallel.
manage_agent
Monitor, control, and retrieve results from spawned agents.