Provides expert knowledge for AWS Amplify Gen 2 development, including TypeScript integration, serverless architecture patterns, and AWS CDK implementation guidance.
Integrates with Neo4j graph database for expert relationship modeling, enabling network exploration, team formation, and relationship queries based on expertise connections.
Provides capabilities for discovering React experts and accessing specialized knowledge for React-based application development.
Offers specialized expertise for TypeScript development, emphasizing type safety, best practices, and integration with AWS services.
Expert Registry MCP Server
Last Updated: 2025-06-30
A high-performance MCP server for expert discovery, registration, and context injection built with FastMCP v2, featuring vector and graph database integration for enhanced semantic search and relationship modeling.
Features
- 🚀 High Performance: Multi-layer caching with vector indices for sub-millisecond queries
- 📁 File-Based Updates: Hot reload on registry/context file changes
- 🔍 Semantic Search: Vector database integration for meaning-based expert discovery
- 🔗 Relationship Modeling: Graph database for expert networks and team formation
- 💉 Context Injection: AI-powered prompt enhancement with expert knowledge
- 📊 Analytics: Performance tracking with collaborative filtering
- 🧠 Hybrid Discovery: Combined vector similarity and graph connectivity scoring
- 🐍 Python-First: Built with FastMCP v2 for clean, Pythonic code
Installation
Docker (Recommended for Production)
The easiest way to run the Expert Registry MCP server is using Docker:
Features:
- 🐳 Single container service for multiple MCP clients
- 📦 Expert contexts and registry mapped to host for easy editing
- 🔄 Hot reload support when files change on host
- 🌐 SSE transport for client connections
- 🗄️ Includes Neo4j database setup
- 🚀 Production-ready with health checks
See DOCKER.md for complete deployment guide.
Local Development
Using uv (recommended):
Using pip:
Database Setup
Vector Database (ChromaDB - Embedded)
Graph Database (Neo4j)
Quick Start
- Set up your expert system directory structure:
- Configure environment:
- Run the server:
Claude Desktop Configuration
Add to your Claude Desktop configuration:
Usage Examples
Basic Expert Discovery
Context Injection
Performance Tracking
Available Tools
Registry Management
expert_registry_list
- List experts with filteringexpert_registry_get
- Get expert detailsexpert_registry_search
- Search experts by query
Expert Selection
expert_detect_technologies
- Detect project technologiesexpert_select_optimal
- Select best expert for taskexpert_assess_capability
- Assess expert capabilityexpert_smart_discover
- AI-powered hybrid search (vector + graph)
Semantic Search
expert_semantic_search
- Search using natural languageexpert_find_similar
- Find similar experts
Graph Operations
expert_explore_network
- Explore expert relationshipsexpert_find_combinations
- Find complementary expert teams
Context Operations
expert_load_context
- Load expert knowledgeexpert_inject_context
- Enhance prompts with expertise
Analytics
expert_track_usage
- Record expert performanceexpert_get_analytics
- Get performance metrics
Expert Registry Format
Expert Context Format
Expert context files are markdown documents in expert-contexts/
:
Development
Setup Development Environment
Run Tests
Code Quality
Architecture
Multi-Layer Caching
- Registry Cache: 24-hour TTL for expert definitions
- Vector Cache: Embeddings cached until expert updates
- Graph Cache: Relationship queries cached for 10 minutes
- Selection Cache: 5-minute TTL for technology detection
- Context Cache: LRU cache for expert contexts (50 entries)
Database Integration
- ChromaDB: Embedded vector database for semantic search
- Multiple collections for different embedding types
- Automatic embedding generation with sentence-transformers
- Neo4j: Graph database for relationship modeling
- Expert-Technology-Task relationships
- Team synergy calculations
- Evolution tracking
Performance Features
- Vector Indices: Annoy indices for ultra-fast similarity search
- Precomputed Combinations: Common expert pairs cached
- Batch Operations: Efficient bulk processing
- Smart Invalidation: Targeted cache updates
File Watching
- Uses
watchdog
for cross-platform file monitoring - Automatic registry reload and database sync
- No server restart required for updates
Troubleshooting
Common Issues
- Expert not found
- Verify expert ID in registry
- Check file paths are correct
- Ensure registry JSON is valid
- Context file missing
- Check expert-contexts directory
- Verify filename matches expert ID
- Ensure .md extension
- Cache not updating
- File watcher may need restart
- Check file permissions
- Verify EXPERT_SYSTEM_PATH
Debug Mode
Enable debug logging:
Advanced Features
Semantic Search
The system uses ChromaDB to enable natural language queries:
Relationship Exploration
Neo4j powers sophisticated relationship queries:
Team Formation
AI-powered team composition:
Contributing
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature
) - Run tests and linting
- Commit your changes (
git commit -m 'Add amazing feature'
) - Push to the branch (
git push origin feature/amazing-feature
) - Open a Pull Request
License
MIT License - see LICENSE file for details
Support
- Documentation: https://github.com/agentience/expert-registry-mcp
- Issues: https://github.com/agentience/expert-registry-mcp/issues
- Discussions: https://github.com/agentience/expert-registry-mcp/discussions
This server cannot be installed
A high-performance MCP server for expert discovery, registration, and context injection, enabling AI-powered expert selection through semantic search and relationship modeling with vector and graph database integration.
Related MCP Servers
- AsecurityAlicenseAqualityAn MCP server integrating Perplexity AI's API to offer advanced search capabilities with support for multiple models and result configuration.Last updated -11JavaScriptMIT License
- -securityFlicense-qualityAn MCP server that integrates real-time web search capabilities into AI assistants using the Exa API, providing both basic and advanced search functionality with formatted markdown results.Last updated -119Python
- -securityAlicense-qualityAn MCP server implementation that standardizes how AI applications access tools and context, providing a central hub that manages tool discovery, execution, and context management with a simplified configuration system.Last updated -9PythonMIT License
- -securityFlicense-qualityAn open-source server implementing the Model Context Protocol (MCP) that enables capturing insights from AI sessions and transforming them into persistent, searchable knowledge accessible across tools.Last updated -TypeScript