hivemind vs teamai-cli
Activeloop's shared brain for agent TEAMS: traces from Claude Code, Codex, Cursor & co become reusable skills every teammate's agent can execute — cloud-backed, 25% cheaper on LoCoMo. — versus — Tencent's team-level agent config manager: skills, rules, agents, hooks, MCP and env live in a shared git repo and sync into Claude Code, Codex, Cursor, CodeBuddy, OpenCode and more on every session.
Both target shared team knowledge for agents — hivemind turns traces into reusable skills for a team, teamai's beta learnings/teamwiki layer does the same on top of its config sync.
| hivemind | teamai-cli | |
|---|---|---|
| Stars | 1.6k | 3.6k |
| Forks | 105 | 230 |
| Language | TypeScript | TypeScript |
| License | Apache-2.0 | NOASSERTION |
| Last activity | yesterday | yesterday |
| Topics | memory, coding | skills, coding |
| Curated connections | 4 | 5 |
hivemind — the curator's take
The pitch is the org-level version of agent memory: one engineer's agent figures out the tricky migration on Monday, every agent on the team executes the pattern Tuesday. Auto-learning from traces across seven agent hosts, with real LoCoMo receipts (25% cheaper, 1.7x fewer tokens vs no shared memory). Reach for it when the unit of learning is the team, not the seat. NOT local-first: cloud-backed on Deeplake is the architecture AND the business model (Activeloop, YC) — traces of your engineers' sessions leave the machine, so clear it with whoever owns your IP policy before the whole team wires in.
teamai-cli — the curator's take
The right tool when the unit is a team, not a developer: an admin curates one 'shared-experience' repo, `teamai init <repo>` installs it per project or per user, and every agent session auto-pulls updates. Coverage is broad (Claude Code, Codex, Cursor, CodeBuddy, WorkBuddy, OpenCode, OpenClaw, Hermes…) and the beta Context/Improvement layers (learnings, codebase graph, sessions, dashboard) push toward a company-wide agent memory. Solo developers should look at apm or skillkit instead — the git-repo-plus-write-access model is overhead without teammates. Tencent-first ecosystem (CodeBuddy, CNB, TGit) — check the matrix for your harness before committing.