Destrayon/Connapse
by Various
Open-source, self-hosted knowledge backend for AI agents — hybrid search (vector + keyword), MCP server, 5 connectors, Docker-ready
MCP
Destrayon/Connapse
Added 1 June 2026
Overview
Destrayon/Connapse is an open-source, self-hosted knowledge backend for AI agents. It provides hybrid search combining vector and keyword methods, includes an MCP server, and comes with five connectors. The project is Docker-ready and written in C#.
Best for
Best for
Developers seeking a lightweight, self-hosted knowledge backend for AI agents with hybrid search capabilities
Use cases
- Deploy a self-hosted knowledge base for AI agents with hybrid search
- Integrate with MCP-compatible agents for retrieval-augmented generation
- Run a private document retrieval system using Docker
How to use
Install
docker-compose up -d Tools exposed
container_createcontainer_listcontainer_deletecontainer_statsupload_filebulk_uploadlist_filesget_documentdelete_filebulk_deletesearch_knowledge
Tested with
Claude Desktop, Claude Code, Cursor, VS Code, ChatGPT
Example client config
{\n "mcpServers": {\n "connapse": {\n "transport": "streamable-http",\n "url": "http://localhost:5001/mcp",\n "headers": {\n "X-Agent-Api-Key": "YOUR_API_KEY"\n }\n }\n }\n} Notes
Destrayon/Connapse is an open-source, self-hosted knowledge backend for AI agents. It provides hybrid search combining vector and keyword methods, includes an MCP server, and comes with five connectors. The project is Docker-ready and written in C#.
13 stars on GitHub. Last updated 2026-05-30. Licensed MIT.
Use cases
- Deploy a self-hosted knowledge base for AI agents with hybrid search
- Integrate with MCP-compatible agents for retrieval-augmented generation
- Run a private document retrieval system using Docker
Pros
- Self-hosted ensures data privacy and control
- Hybrid search (vector + keyword) improves retrieval accuracy
- Docker-ready simplifies deployment and scaling
Cons
- Very small community with only 13 GitHub stars, limited support
- C# codebase may reduce contributor pool compared to Python or JavaScript
- Unclear documentation and feature completeness given early-stage project
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Self-hosted ensures data privacy and control
- Hybrid search (vector + keyword) improves retrieval accuracy
- Docker-ready simplifies deployment and scaling
Cons
- Very small community with only 13 GitHub stars, limited support
- C# codebase may reduce contributor pool compared to Python or JavaScript
- Unclear documentation and feature completeness given early-stage project
Open-source & AI alternatives
Swap-in tools that solve the same job. Weigh the trade-offs before you commit.
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
firecrawl/firecrawl-mcp-server
Various
🔥 Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.
exa-labs/exa-mcp-server
Various
Exa MCP for web search and web crawling!
upstash/context7
Various
Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors
microsoft/markitdown
Various
Python tool for converting files and office documents to Markdown.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.
