MCP ServersDocsearch Mcp
Created at 21 days ago
by PatrickKoss
ripgrep for docs but via MCP
Categories
Tags
docsearch
document-search
mcp
What is Docsearch MCP?
Docsearch MCP is a local-first document search and indexing system that provides hybrid semantic and keyword search across local files (including PDFs) and Confluence pages through the Model Context Protocol (MCP). It is designed to assist AI tools in accessing documentation, codebases, and research materials.
How to use Docsearch MCP?
To use Docsearch MCP, you can install it via npm or run it using Docker. After installation, you can index your documents and search through them using the command-line interface (CLI).
Key features of Docsearch MCP?
- Hybrid Search: Combines full-text search with vector similarity for optimal results.
- Multi-Source Indexing: Index local files and Confluence spaces.
- PDF Support: Extract and search text from PDF documents.
- Image Search: AI-powered image description and search.
- Database Flexibility: Supports SQLite and PostgreSQL.
- Real-time Updates: Automatic re-indexing with file watching.
- Multiple Output Formats: Supports text, JSON, and YAML outputs.
Use cases of Docsearch MCP?
- Searching through code documentation.
- Indexing and searching academic papers.
- Assisting AI tools in retrieving relevant documentation.
FAQ from Docsearch MCP?
- Can Docsearch MCP index all file types?
Yes, it supports various file types including code files, PDFs, and images.
- Is Docsearch MCP free to use?
Yes, it is open-source and free to use.
- How does the hybrid search work?
It combines traditional keyword search with semantic search using vector embeddings.
View More