OpenDots vs qm
CopilotKit template for always-on AI coworkers ('Dots'): each gets a role, tools and its own computer, works in chat, calls and Slack, and saves approved drafts into editable Spaces. — 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 give team agents scoped workspaces reachable from Slack and web. QM swaps coding harnesses underneath; OpenDots gives each Dot its own computer and a shared document workspace with approval gates.
| OpenDots | qm | |
|---|---|---|
| Stars | 6.4k | 16k |
| Forks | 940 | 1.9k |
| Language | TypeScript | TypeScript |
| License | MIT | MIT |
| Last activity | yesterday | yesterday |
| Topics | agents, orchestration | agents, orchestration |
| Curated connections | 5 | 9 |
OpenDots — the curator's take
Use it as a starting point for a self-hosted agent workspace where specialist agents share a document library with humans — per-Dot permissions, approve-before-run cards for MCP tools, and an optional persistent browser/terminal per Dot. NOT a hosted product or a plain library: alpha, you run the infrastructure, and conversations need CopilotKit Intelligence (hosted, local Docker eval, or a licensed self-host) — a real dependency to weigh. Want only the governed computer per agent? Take OpenBot.
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.