spark-memory 0.1.2

High-performance MCP memory server powered by Redis Stack and LangGraph

Spark Memory ✨

High-performance MCP (Model Context Protocol) memory server powered by Redis Stack and LangGraph.

Features

• ⚡ Lightning Fast: Redis Stack-based with millisecond response times
• 🧠 Intelligent Memory: Vector search and semantic consolidation
• 🔒 Enterprise Security: Field-level encryption, RBAC, audit logging
• 📈 Scalable: Distributed environment support
• 🎯 Easy Deployment: One-line execution via uvx

Installation

Requirements

• Python 3.11+
• Redis Stack 7.2.0+ (with JSON, Search, TimeSeries modules)

Install via pip

pip install spark-memory

Install Redis Stack

macOS:

brew tap redis-stack/redis-stack
brew install redis-stack
brew services start redis-stack

Ubuntu/Debian:

curl -fsSL https://packages.redis.io/gpg | sudo gpg --dearmor -o /usr/share/keyrings/redis-archive-keyring.gpg
echo "deb [signed-by=/usr/share/keyrings/redis-archive-keyring.gpg] https://packages.redis.io/deb $(lsb_release -cs) main" | sudo tee /etc/apt/sources.list.d/redis.list
sudo apt-get update
sudo apt-get install redis-stack

Docker:

docker run -d --name redis-stack -p 6379:6379 -p 8001:8001 redis/redis-stack:latest

Quick Start

1. Run as MCP Server

# Using uvx (recommended)
uvx spark-memory

# Or with Python
python -m spark_memory

2. Configure with Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "spark-memory": {
      "command": "uvx",
      "args": ["spark-memory"]
    }
  }
}

3. Use in Claude

Once configured, you can use these commands in Claude:

• m_memory: Save, search, and manage memories
• m_state: Checkpoint and state management
• m_admin: System administration
• m_assistant: Natural language memory commands

Architecture

Spark Memory uses a layered architecture:

1. MCP Server Layer: FastMCP server exposing tools via Model Context Protocol
2. Memory Engine Layer: Core business logic for memory operations
3. Redis Client Layer: Redis Stack wrapper with connection pooling
4. Security Layer: Encryption, access control, and audit logging

Memory Operations

Save Memory

# In Claude, you can say:
# "Save this conversation about Python optimization"
# "Remember that the meeting is at 3pm tomorrow"

Search Memory

# "What did we discuss about Redis?"
# "Find all memories from last week"

State Management

# "Create a checkpoint for the current project"
# "Restore the previous state"

Configuration

Environment variables:

• REDIS_HOST: Redis server host (default: localhost)
• REDIS_PORT: Redis server port (default: 6379)
• REDIS_PASSWORD: Redis password (optional)
• LOG_LEVEL: Logging level (default: INFO)

Development

# Clone the repository
git clone https://github.com/Jaesun23/spark-memory
cd spark-memory

# Install development dependencies
pip install -e ".[dev]"

# Run tests
pytest tests/

# Run with debug logging
LOG_LEVEL=DEBUG python -m spark_memory

License

MIT License - see LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions

---

Made with ❤️ by Jason and 1호