AI for Developers
Agent skill (SKILL.md) Open source Apache-2.0 1.6k

Hivemind

Hivemind turns your traces into reusable skills across agents

Install

For this project
$ npx skills add activeloopai/hivemind -a claude-code
For every project
$ npx skills add activeloopai/hivemind -a claude-code -g

Or copy the skill folder into .claude/skills/ (project) or ~/.claude/skills/ (all projects).

For this project
$ npx skills add activeloopai/hivemind -a codex
For every project
$ npx skills add activeloopai/hivemind -a codex -g

Or copy the skill folder into .agents/skills/ (project) or ~/.agents/skills/ (all projects).

For this project
$ npx skills add activeloopai/hivemind -a openclaw
For every project
$ npx skills add activeloopai/hivemind -a openclaw -g

Or copy the skill folder into skills/ (project) or ~/.openclaw/skills/ (all projects).

See the repository README for installation instructions.

When to use

Use it when a team runs several agents and wants what one learns to benefit the rest, and you're comfortable with traces being stored.

Hivemind is a cloud-backed shared memory for a team's agents, for Claude Code, OpenClaw, Codex, Cursor, Hermes, Pi and Claude Cowork (alpha). It captures every session's prompts, tool calls and responses as structured traces in Deeplake, then mines them for repeated patterns and codifies those into reusable SKILL.md files available to every agent on the team. The README's example is one engineer's agent working out a tricky migration on Monday, and every agent on the team being able to run the pattern on Tuesday.

On the public LoCoMo long-context memory benchmark it reports 25% lower cost, 1.7 times fewer tokens and 31% fewer turns than running without shared memory, which is the vendor's own measurement. Because it stores full session traces in the cloud, check what that means for your code and secrets.

Related skills

All skills