auto-co-meta

auto-co is a Bash loop + Claude CLI that runs a 14-agent AI company autonomously. The whole thing is a bash loop: while true; do claude -p "$PROMPT" --dangerously-skip-permissions sleep 120 done Each cycle, the agents read a shared `memories/consensus.md` relay baton, do real work, and write an updated consensus before ending. The relay baton is what keeps the company coherent across unlimited cycles. After 22 cycles (~3 weeks, $36 total AI cost): - Next.js landing page live on Railway - Live demo dashboard at /demo - Waitlist email capture (Supabase) - GitHub star counter - DEV.to article published (by the agents) - This HN post drafted (by the agents) The 14 agents are modeled on real experts: Jeff Bezos (CEO), DHH (Engineering), Seth Godin (Marketing), Charlie Munger (Critic). Munger's rule: before any major decision, he r

Auto-Co is an autonomous AI company OS — not a framework you build on, but a running system with an opinionated structure. Architecture: - 14 agents with expert personas (CEO/Bezos, CTO/Vogels, CFO/Campbell, Critic/Munger...) - Bash loop + Claude Code CLI — no custom inference, no vector stores - Shared markdown consensus file as the cross-cycle relay baton - Human escalation via Telegram for true blockers only (2 escalations in 12 cycles) - Every cycle must produce artifacts: code, deployments, docs The repo IS the live company. It built its own landing page, README, Docker stack, GitHub release, and community posts — all autonomously across 12 cycles of self-improvement. What makes it different from AutoGen/CrewAI/LangGraph: those are building blocks. Auto-Co is the building. The decision hierarchy, safety guardrails, and convergence ru