-
-
Notifications
You must be signed in to change notification settings - Fork 110
Home
Henry edited this page Aug 22, 2025
·
20 revisions
Welcome to the MCP Memory Service Wiki! This is your comprehensive guide to leveraging AI-powered memory management in your daily workflows.
MCP Memory Service is a powerful tool that extends AI assistants with persistent, searchable memory capabilities. It allows you to:
- Store important information, decisions, and learnings
- Retrieve relevant context from past conversations
- Build a personal knowledge base that grows with you
- Organize information with tags and natural language search
# Store a memory
\"Remember that our deployment process uses GitHub Actions with staging first\"
# Retrieve memories
\"What do you know about our deployment process?\"
# Search by time
\"What did we discuss yesterday about the API changes?\"
# Search by tags
\"Show me all memories tagged with 'python' and 'debugging'\"
Morning: \"Recall what I was working on yesterday\"
During work: \"Remember that the API endpoint /users/profile has a rate limit issue\"
End of day: \"Summarize today's progress and key decisions\"
\"Remember this solution for handling async operations in Python\"
\"Store this regex pattern for parsing log files: [pattern]\"
\"What debugging techniques have I used for memory leaks?\"
\"Store meeting notes: Product roadmap discussion - Q2 priorities include...\"
\"What were the action items from last week's planning meeting?\"
\"Remember John is the point person for the authentication service\"
\"Remember my preferred VSCode settings for Python development\"
\"Store this bash alias for quick database connections\"
\"What's my usual process for reviewing PRs?\"
Create periodic memory maintenance routines:
Weekly Review Prompt:
\"Every Friday, run memory maintenance:
1. Review and tag memories from this week
2. Consolidate related memories
3. Archive completed project memories
4. Generate a weekly summary\"
Daily Standup Helper:
\"Each morning:
1. Recall yesterday's tasks and progress
2. Check for blocked items
3. Review today's priorities
4. Store the daily plan\"
Project Context:
\"For all conversations about [project-name]:
- Automatically store technical decisions
- Remember configuration changes
- Track dependencies and versions
- Note team member responsibilities\"
Learning Capture:
\"When I say 'TIL' (Today I Learned):
- Store the learning with tag 'til'
- Add relevant technology tags
- Include the date
- Link to related memories\"
Error Solutions:
\"When we solve an error or bug:
- Store the error message
- Remember the solution
- Tag with 'debugging' and the technology
- Include any relevant code snippets\"
🚀 REVOLUTIONARY UPDATE: Claude Code Memory Awareness
- Phase 2 Complete: Dynamic memory loading during conversations with real-time topic detection
- Conversation Intelligence: NLP engine analyzing topics, entities, intent, and code context
- Cross-Session Intelligence: Conversation threading and progress continuity across sessions
- 100% Test Coverage: Production-ready with comprehensive integration testing
- <500ms Response Time: Lightning-fast context updates that don't interrupt your flow
Transform your development workflow with intelligent memory that:
- Remembers your project context across all coding sessions
- Learns from conversations and provides dynamic context updates
- Connects related discussions across different sessions
- Provides perfect timing for relevant information injection
- 🧠 Claude Code Memory Awareness Guide - NEW! Complete guide to intelligent development with Phase 1 & 2
- ⚡ Claude Code Quick Reference - NEW! Essential commands, configuration, and troubleshooting
- 🔍 Debugging Case Study - How we went from crisis to v5.0.4
- 🧠 CLAUDE.md Memory Pattern - Our innovative 3-layer context system
- 🧠 Claude Code Memory Awareness Guide - MUST READ! Complete workflow guide for intelligent development with memory-aware coding
- ⚡ Claude Code Quick Reference - START HERE! Essential commands, setup, and troubleshooting for immediate productivity
- 🔍 Debugging Case Study - Real-world debugging journey from production failure to major improvements
- 🧠 CLAUDE.md Memory Pattern - Revolutionary context management for AI development (65% token reduction!)
- Memory Cleanup Example - Learn how to maintain a clean, organized memory database
- Tag System Migration and Management - Comprehensive guide to organizing memories with tags
- Building a React Dashboard - Create a visual interface for your memories
- Performance Optimization - Make your memory service lightning fast
- 🔧 Troubleshooting Guide - NEW! Comprehensive solutions for deployment, database, and service issues
- Timestamp Precision Fix - How natural language time queries work
- Debugging MCP Protocol - Troubleshooting guide for common issues
- Query Time Tracking - Implementation case study
- Memory Re-tagging System - Three-phase approach to organization
- Integration Patterns - Connect with other tools and services
- Best Practices - Tips for effective memory management
- Ngrok Connectivity Guide - Secure external access for remote clients
- Store code snippets and solutions
- Track architectural decisions
- Remember debugging sessions
- Document API changes and migrations
- Capture key insights from articles
- Build a personal knowledge base
- Track learning progress
- Connect related concepts
- Share team knowledge
- Document decisions and rationale
- Track project evolution
- Maintain institutional memory
- Remember preferences and settings
- Track habits and routines
- Store quick references
- Build personal documentation
Thanks to community feedback from Joe Esposito, we completely simplified Docker in v5.0.4!
🤖 MCP Mode (for Claude Desktop, VS Code):
docker-compose up -d🌐 HTTP Mode (for REST API, Web Dashboard):
docker-compose -f docker-compose.http.yml up -d
curl http://localhost:8000/api/health # It just works!That's it! No more confusion about which of 4 files to use.
- ✅ Fixed missing
run_server.py - ✅ Corrected Python paths
- ✅ Pre-download models (no more runtime failures)
- ✅ Smart unified entrypoint
- ✅ 60% complexity reduction
Joe Esposito discovered critical Docker issues and helped us completely overhaul the setup. His feedback led to the v5.0.4 Docker revolution - from 4 confusing configs to 2 crystal-clear options!
"Thanks for your quick reply Claude, lol... The good news is that all the dashboards are working now!" - Joe
This is exactly the kind of real-world feedback that makes MCP Memory Service better for everyone.
Claude Code Memory Awareness (v6.1.0) in Production:
- 100% test pass rate (15/15 comprehensive integration tests)
- <500ms response time for dynamic context updates
- Conversation intelligence with 15+ technical topic categories
- Cross-session threading connecting related discussions over time
- Zero cognitive load context updates during active coding
CLAUDE.md Memory Pattern Results:
- 750+ memories stored and actively used
- 65% token reduction in Claude Code sessions
- 96.7% faster context setup (15min → 30sec)
- 100% knowledge retention across sessions
Docker Platform Stability:
- 60% less complexity (4 configs → 2)
- 73% faster startup with pre-downloaded models
- Zero confusion about deployment modes
Real metrics from real usage in production environments!
Link related memories for context:
\"Create a memory chain for 'Project X Migration':
- Initial planning decisions
- Technical challenges encountered
- Solutions implemented
- Lessons learned\"
Set up automatic memory retrieval:
\"When discussing AWS services, automatically recall:
- Our AWS configuration standards
- Common troubleshooting steps
- Cost optimization strategies\"
Create reusable formats:
\"Bug Report Template:
- Error message
- Steps to reproduce
- Environment details
- Solution applied
- Prevention measures\"
We welcome contributions from the community! Here's how you can help:
- Share Your Workflows: Document how you use the memory service in your daily work
- Use Case Examples: Add real-world examples of problems solved
- Tips & Tricks: Share productivity hacks and best practices
- Integration Guides: Document how you've connected the memory service with other tools
- Technical Guides: Write about implementation details and optimizations
- Troubleshooting: Document issues you've encountered and solutions
- Feature Tutorials: Create guides for new features you've built
- Code Examples: Share useful scripts and utilities
- Be Practical: Focus on real-world usage and examples
- Include Code: Provide runnable examples where possible
- Document Results: Show before/after states and performance metrics
- Stay Organized: Follow the existing structure and naming conventions
- Test Everything: Ensure your examples work as described
- Fork the repository
- Create your wiki page following our structure
- Include a clear table of contents
- Add your page to the relevant section in this home page
- Submit a pull request with a description of your addition
- Integration with popular development tools
- Domain-specific memory templates (DevOps, Data Science, etc.)
- Automation scripts and workflows
- Performance optimization techniques
- Multi-user collaboration patterns
- Mobile and cross-platform usage guides
Stay tuned for upcoming features:
- Visual memory mapping
- Advanced analytics and insights
- Team collaboration features
- API integrations
- Mobile applications
- Voice interaction capabilities
- 🔧 Troubleshooting: Start with our comprehensive Troubleshooting Guide for common deployment and service issues
- Issues: Report bugs and request features on GitHub Issues
- Discussions: Join the community in GitHub Discussions
- Documentation: Check the README for setup instructions
Remember: The MCP Memory Service is more than a tool—it's your personal knowledge companion that grows smarter with every interaction. Start small, be consistent, and watch your digital memory become an invaluable asset in your daily work.
Happy memorizing! 🎉