ClawMem provides an on-device hybrid memory system for AI agents, combining BM25 and vector search with LLM-powered query expansion, reranking, and graph-based causal traversal. It exposes hooks for automatic context surfacing and MCP tools for explicit retrieval and lifecycle actions (pin, snooze, forget).
Use ClawMem when you need reliable, local-first memory for agents: surfacing relevant context into prompts, answering cross-session 'why/when' questions, building temporal/semantic graphs after bulk ingestion, or managing memory lifecycle policies for long-lived decisions. Prefer hooks for per-turn context and MCP tools for escalated retrieval, debugging, or maintenance tasks.
ClawMem targets OpenClaw/Hermes and Claude Code environments; it integrates with MCP tools and can be used alongside agent runtimes that support local CLI or REST tool calls.
This skill has not been reviewed by our automated audit pipeline yet.
