Hivemind is a shared memory and skill system for AI coding agents. It captures agent sessions as structured traces, including prompts, tool calls, responses, and subagent activity. The system mines those traces for repeated patterns and turns them into reusable skills that can be shared across a team. It supports agents such as Claude Code, OpenClaw, Codex, Cursor, Hermes, and pi. Hivemind uses hybrid lexical and semantic retrieval so agents can recall prior decisions, solved bugs, migrations, and team-specific workflows. It is useful for engineering teams that want agent behavior to improve over time instead of starting from zero in every session.
Features
- Shared memory across coding agents
- Structured trace capture
- Reusable skill generation
- Hybrid lexical and semantic search
- Claude Code, Codex, Cursor, and OpenClaw support
- Workspace-based team knowledge propagation
