MCP Knowledge Service
Enables semantic search and management of development knowledge including global rules, project documentation, and references through vector-based search using libSQL. Features Tailscale-secured access control and tools for searching, browsing, and organizing development resources across multiple channels.
README
MCP Knowledge Service
MCP-based knowledge and rules suite for Tailscale networks with semantic search via libSQL vectors.
Features
- MCP Tools:
rules.search,project.search,refs.list, and more - Semantic Search: Vector-based search using libSQL with OpenAI embeddings
- Multi-Channel: Support for multiple MCP channels (rules, projects, refs)
- Tailscale Security: Identity-based access control via Tailscale Serve
- Vector Database: libSQL with native vector support for ANN search
Quick Start
-
Setup Environment:
./scripts/setup.sh -
Configure Environment: Edit
.envwith your configuration:LIBSQL_URL=file:./data/knowledge.db OPENAI_API_KEY=your-openai-api-key -
Development:
npm run dev # Start development server npm run build # Build for production npm test # Run tests npm run lint # Lint code
Architecture
src/mcp/- MCP server and tool implementationssrc/db/- Database schema, connections, and migrationssrc/http/- REST API endpoints for ingestionsrc/auth/- Tailscale identity and access controlsrc/utils/- Shared utilities and helpers
MCP Tools
Rules Service
rules.search(q, k?, tags?)- Search global development rulesrules.get(id)- Get specific rule by IDrules.tags()- List all available rule tags
Project Service
project.search(project, q, k?)- Search within project docsproject.browse(project, path?)- Browse project structureproject.contextPack(project, facets?)- Get curated context bundle
References Service
refs.list(tags?, limit?)- List references with optional tag filterrefs.add(title, url, note?, tags_csv?)- Add new referencerefs.findByTag(tag)- Find references by specific tag
Database Schema
The service uses libSQL with vector support:
rules_global- Global AI development rules with embeddingsproject_docs- Project-specific documentation with embeddingsrefs- Quick reference links and documentationaccess_tiers- User access control and permissionsaudit_log- Query audit trail and metrics
Development Status
See TODO.md for current development phases and tasks.
Related
- Memory subsystem development:
/home/ubuntu/mem - Design documentation:
docs/memory-design.md
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.