Skip to main content

Amazon Q CLI용 대화 기억 및 세션 관리 MCP 서버 - ChromaDB 기반 실시간 동기화

Project description

Q Memory MCP Server

A Model Context Protocol (MCP) server that provides conversation memory and session management for Amazon Q CLI.

🎯 Key Features

  • Automatic Conversation Saving: Real-time sync of Q CLI conversations to ChromaDB
  • Session Management: Organize conversations by topics/sessions
  • Context Restoration: Resume previous conversations with full context memory
  • Semantic Search: Search through conversation history using natural language

🚀 Quick Start

1. Installation

cd /path/to/q-mem-mcp-server
pip install -e .

2. MCP Configuration

Add to ~/.config/q/mcp_config.json:

{
  "mcpServers": {
    "q-mem": {
      "command": "python",
      "args": ["-m", "q_mem_mcp.server"],
      "env": {}
    }
  }
}

3. Usage

# Start Q CLI
q chat

# Start a new session
start_session(description="Python Learning")

# Chat normally (automatically saved)
# ... have conversations ...

# List sessions
list_sessions()

# Resume session (loads full context)
resume_session(session_id="Python_Learning_0709_1234")

🛠️ Available Commands

Command Description
start_session(description) Start a new session
list_sessions() List all sessions
resume_session(session_id) Resume session with full context
search_memory_by_session_id(session_id, query) Search previous conversations
delete_session(session_id, confirm=true) Delete a session
get_storage_stats() Check storage status

🔧 Technology Stack

  • ChromaDB: Vector database for conversation storage and search
  • SQLite WAL: Real-time sync with Q CLI database
  • Sentence Transformers: Semantic search embeddings
  • MCP Protocol: Communication with Amazon Q

📁 Data Storage

  • ChromaDB: ~/.Q_mem/chroma_db/
  • Sync State: ~/.Q_mem/sync_state.json
  • Logs: ~/.Q_mem/q_mem.log

🔄 Auto-Sync Features

Q CLI conversations are automatically saved to ChromaDB in real-time:

  • Real-time Detection: Checks for new conversations every 2 seconds
  • Partial Failure Handling: Saves successful conversations even if some fail
  • Auto Recovery: Automatically recovers from consecutive failures
  • State Restoration: Restores sync state after restart

💡 Usage Tips

  1. Session Organization: Separate conversations by topics
  2. Semantic Search: Use natural language rather than exact keywords
  3. Context Utilization: Use resume_session for complete conversation restoration
  4. Regular Cleanup: Delete unnecessary sessions with delete_session

🐛 Troubleshooting

Q CLI Sync Issues

# Set environment variable
export Q_CLI_DB_PATH="/path/to/q/cli/database"

# Or check auto-detection
get_storage_stats()

ChromaDB Errors

# Recreate database
rm -rf ~/.Q_mem/chroma_db/
# Run start_session again

Memory Issues

# Clean up old sessions
cleanup_old_sessions(days=30, confirm=true)

📄 License

MIT License

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

q_mem_mcp_server-1.0.0.tar.gz (23.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

q_mem_mcp_server-1.0.0-py3-none-any.whl (26.3 kB view details)

Uploaded Python 3

File details

Details for the file q_mem_mcp_server-1.0.0.tar.gz.

File metadata

  • Download URL: q_mem_mcp_server-1.0.0.tar.gz
  • Upload date:
  • Size: 23.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for q_mem_mcp_server-1.0.0.tar.gz
Algorithm Hash digest
SHA256 8f0695ac4076814f7b1e39b98ea3523f8e9e0e85780bf1770789fddbc80d1b48
MD5 9c75e84dda2692070e3c7a17bd992a4b
BLAKE2b-256 95f8fdee1f7d8eda1e42709b62cf625bf04b0d8e612aefbbb8092fb0c770597b

See more details on using hashes here.

File details

Details for the file q_mem_mcp_server-1.0.0-py3-none-any.whl.

File metadata

File hashes

Hashes for q_mem_mcp_server-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 632710d68ea21ddb313ee0a00a9875806b0cd3f1bb15cb41baacbf0f94eae252
MD5 e750f45e8ff141600371420e46c03b64
BLAKE2b-256 2b5450b2c15ff41b490a718c1bcb4266bf240436009f648fbcb5ce68610b597a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page