hivemind
View on GitHubHivemind turns your traces into reusable skills across agents
Hivemind captures AI agent session traces, retrieves relevant memories, and turns repeated patterns into reusable skills shared across agents and teammates. It integrates with Claude Code, Codex, Cursor, OpenClaw, Hermes, and pi, with cloud-backed Deeplake storage.
Use Cases
Share prior agent session knowledge across a teamCapture and search agent prompts, tool calls, and responsesTurn repeated workflows into reusable agent skillsRecall relevant traces and skills in coding agentsSummarize agent sessions into wiki pages
Built With
- Language
- TypeScript
- Frameworks
- Deeplake · Model Context Protocol · Anthropic SDK
Tags
agent memory · shared memory · long-term memory · trace capture · skill generation · semantic search · hybrid retrieval · multi-agent · Claude Code · Codex · Cursor · OpenClaw · MCP · team knowledge