exxperts vs qm
Local-first agentic runtime with persistent AI rooms and approval-gated memory: every memory write needs your OK; rooms, KB and artifacts are plain files on disk. — versus — Multiplayer agent harness for startups: every employee gets a scoped workspace — memory, files, keychain, crons, sandbox — in Slack and web, with Pi/OpenCode/Codex/Claude Code swappable underneath.
Both scope agents into rooms with isolated, persistent memory. Exxperts is local-first single-user with approval-gated memory writes; qm is the org-wide version — Postgres core, Slack/web identity, admin-set security postures.
| exxperts | qm | |
|---|---|---|
| Stars | 268 | 7.2k |
| Forks | 28 | 758 |
| Language | TypeScript | TypeScript |
| License | Apache-2.0 | MIT |
| Last activity | 4 days ago | 2 days ago |
| Topics | memory, agents, local | agents, orchestration |
| Curated connections | 4 | 3 |
exxperts — the curator's take
Use it when you want a personal AI that remembers you but on your terms — every memory write passes an approval gate, and all state is auditable files under ~/.exxperts. Two surfaces (web app + CLI/TUI) share the same rooms. NOT a chat-over-docs RAG frontend (that's Open WebUI territory) and not a coding agent — rooms never get an unrestricted shell. Single-user, local-only by design; and mind the PolyForm Noncommercial license: commercial use needs a separate deal with Exxeta.
qm — the curator's take
The org-level answer to "give everyone at the company an agent": per-person and per-channel scopes keep memory, credentials and files isolated, and the harness underneath (Pi, OpenCode, Codex, Claude Code) is swappable, so the deployment isn't a bet on one vendor. Security postures (strict/auto/dangerous) plus a hard command-deny policy that survives even Dangerous mode are unusually well thought through. NOT for individuals — this is Postgres + sandbox + deployment-repo infrastructure; a solo dev should run a harness directly. Quirk: contributions are accepted as prose ADRs, not code PRs.