claude-mem: what it is, what problem it solves & why it's gaining traction

What it solves

Claude-Mem provides persistent memory for AI agents, specifically designed for Claude Code, Gemini CLI, and OpenCode. It solves the problem of context loss between sessions, allowing an AI to remember project history, tool usage observations, and key decisions across different conversations.

How it works

The system uses a combination of lifecycle hooks (such as SessionStart and PostToolUse) to automatically capture observations and generate semantic summaries. These are stored in a SQLite database and a Chroma vector database for hybrid semantic and keyword search. A worker service manages an HTTP API and a web viewer UI for real-time memory monitoring.

Who it’s for

Developers using Claude Code, Gemini CLI, or OpenCode who want their AI assistant to maintain continuity of knowledge and project context without manual prompting in every new session.

Highlights

• Persistent Memory: Context survives across sessions automatically.
• Token-Efficient Search: A 3-layer workflow (search $\rightarrow$ timeline $\rightarrow$ get_observations) reduces token costs by filtering results before fetching full details.
• Hybrid Search: Combines full-text and vector search via Chroma DB for intelligent retrieval.
• Web Viewer UI: Real-time memory stream available at localhost:37777.
• Privacy Controls: Supports <private> tags to exclude sensitive content from storage.
• Beta Features: Includes an experimental "Endless Mode" for extended sessions.