enhanced -memory

Enhanced MCP Memory provides an intelligent memory and task management server designed to help AI assistants track memories, extract tasks from conversations and code, and manage projects with a knowledge-graph backbone. It exposes a rich set of core memory tools (get_memory_context, create_task, get_tasks, get_project_summary), a sequential thinking engine for structured reasoning, and project-convention learning to automatically detect project patterns and suggest correct commands. Use uvx to run the server with the included enhanced-mcp-memory entry point, and interact with the exposed tools to search memories, auto-extract tasks, organize projects, and monitor performance.

Prerequisites:

• Python 3.8+
• uvx (for one-command installation) or Python environment with dependencies

Option A: Quick uvx install (recommended)

1. Ensure uvx is installed on your system
2. Run: uvx enhanced-mcp-memory

Option B: Manual installation

1. Clone the repository: git clone https://github.com/cbunting99/enhanced-mcp-memory.git
2. Change into the directory: cd enhanced-mcp-memory
3. Install Python dependencies: pip install -r requirements.txt
4. Run the server directly: python mcp_server_enhanced.py

Option C: Development setup

1. Clone and install dependencies as above
2. Run directly for development: python mcp_server_enhanced.py

Environment variables (LOG_LEVEL, MAX_MEMORY_ITEMS, ENABLE_AUTO_CLEANUP) control logging verbosity, memory capacity, and automatic cleanup behavior. Adjust MAX_MEMORY_ITEMS to fit your memory budget. The server uses a SQLite database under data/ and stores logs under logs/. If you see performance issues, review health_check() and get_performance_stats() outputs, then consider running cleanup_old_data(days_old) or optimize_memories(). For Windows users, ensure path handling and command formats align with Windows conventions. The configuration shown is for uvx usage; if you deploy locally, switch to the Python invocation with the provided cwd path.