Hivemind Memory

If you're running Claude in a team or org context, this gives every agent instance access to a shared memory layer backed by Activeloop's Deep Lake. It indexes session logs into searchable summaries at ~/.deeplake/memory/, so when you ask Claude to recall something, it checks both its local project memory and the global Hivemind store. The standout feature is "skillify", which mines reusable skills from session logs and lets you pull them into new projects with filters by author or team. There's optional semantic search via embeddings (needs a 600MB download) and org management commands for invites and workspace switching. It's bash-only for filesystem ops, which keeps things simple but means no Python scripts in the memory layer.

GitHub stars: 1.1K
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add activeloopai/hivemind --skill hivemind-memory --agent claude-code

Installs into .claude/skills of the current project.

Files

Hivemind Memory

You have TWO memory sources. ALWAYS check BOTH when the user asks you to recall, remember, or look up ANY information:

Your built-in memory (~/.claude/) — personal per-project notes
Hivemind global memory (~/.deeplake/memory/) — global memory shared across all sessions, users, and agents in the org

Memory Structure

~/.deeplake/memory/
├── index.md                          ← START HERE — table of all sessions
├── summaries/
│   ├── session-abc.md                ← AI-generated wiki summary
│   └── session-xyz.md
└── sessions/
    └── username/
        ├── user_org_ws_slug1.jsonl   ← raw session data
        └── user_org_ws_slug2.jsonl

How to Search

First: Read ~/.deeplake/memory/index.md — quick scan of all sessions with dates, projects, descriptions
If you need details: Read the specific summary at ~/.deeplake/memory/summaries/<session>.md
If you need raw data: Read the session JSONL at ~/.deeplake/memory/sessions/<user>/<file>.jsonl
Keyword search: Grep pattern="keyword" path="~/.deeplake/memory"

Do NOT jump straight to reading raw JSONL files. Always start with index.md and summaries.

SKILL.md on GitHub