brigade vs picobot
Self-hosted crew of AI agents on an org chart sharing one long-term memory: delegation, mid-task model switching, 1,000+ app connectors, reachable from terminal, WhatsApp, Telegram or Slack. — versus — A self-hosted personal AI agent in a single ~9MB Go binary — persistent memory, 16 tools + MCP, skills, cron/heartbeat, and Telegram/Discord/Slack/WhatsApp channels. Runs on a $5 VPS.
Both are self-hosted personal agents with persistent memory, skills, cron and messaging channels; picobot is one small Go binary, Brigade a multi-agent crew with an org chart.
| brigade | picobot | |
|---|---|---|
| Stars | 11k | 1.3k |
| Forks | 70 | 168 |
| Language | TypeScript | Go |
| License | MIT | MIT |
| Last activity | 2 days ago | 24 days ago |
| Topics | agents, orchestration | agents |
| Curated connections | 3 | 5 |
brigade — the curator's take
For one owner who wants a personal multi-agent assistant they host themselves — reuses your Claude/ChatGPT/Copilot sign-in, approval-gated privileged actions, no telemetry. The README is heavy on vibes (the 'Bloody Benchmark' is a public tunnel, not a benchmark), so judge the code, not the copy. One lean agent on a $5 VPS? picobot is simpler. Company-scale fleets with budgets and governance? paperclip.
picobot — the curator's take
The anti-bloat statement piece: zero dependencies, ~10MB RAM idle, instant cold start — an always-on agent with ranked memory recall, background subagents, a natural-language HEARTBEAT.md cron, and self-authored skills ('create a skill for checking weather' → it writes the markdown), reachable from your phone via four chat channels. Runs on a Raspberry Pi or Termux on an old Android. Any OpenAI-compatible endpoint, including Ollama. Its own README names the target: OpenClaw's power without the 500MB container. NOT a coding harness and single-user by design — this is your personal daemon, not a team platform; complex multi-agent workflows will outgrow it fast, which is rather the point.