Ross ROSS = Recommend OSS · open-source software intelligence for agents

compozy/compozy

An operating system for AI agents. Plug in the agent CLIs you already use (Claude Code, Codex, Gemini CLI, Cursor) and they become a team: they split the work, hand tasks to each other, run automated on jobs and loops, and share one project memory. You steer everything from the browser. observed · 2026-08-28

github.com/compozy/compozy · homepage · Go · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

79/100

  • Activity 99
  • Release rhythm 93
  • Longevity 11

Flags: young

How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 1.0
  • age_days: 158
  • days_rel: 47
  • days_push: 7
  • n_releases_24m: 29

Full methodology

Adoption not part of the score

2686 stars · 168 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

CompozyOS is a self-hosted 'operating system for AI agents' written in Go that wraps existing agent CLIs (Claude Code, Codex, Gemini CLI, OpenClaw, Hermes) into daemon-owned sessions with shared memory, loops, approvals, and permissions. It provides a web UI, CLI, HTTP/SSE, UDS, and MCP surfaces for creating, automating (cron, webhooks, triggers), and supervising agent work locally.

Use cases

  • orchestrate multiple AI agent CLIs as a coordinated team
  • run AI agents on cron schedules or webhooks without a terminal open
  • give coding agents shared project memory across sessions
  • supervise agent runs with approvals and permissions from a browser
  • delegate tasks between agents with handoffs and task leases
  • consolidate agent activity into durable memory automatically
  • run unattended background agent jobs with audit trails

When to choose

  • you already use agent CLIs like Claude Code or Gemini CLI and want them running continuously and collaboratively
  • you need scheduled, event-driven, or webhook-triggered agent automation with supervision and audit history
  • you want a local-first, single-binary agent runtime with file-based memory you can version and inspect

When to avoid

  • you only need a simple chatbot or one-off LLM API calls without orchestration
  • you require a fully stable, production-hardened product - the v0.3 line is explicitly beta
  • you need a managed cloud service rather than a self-hosted daemon on your own machine
  • your agents are not ACP-compatible CLIs or you rely on providers outside its built-in integrations

Facets

application · maturity experimental

agent-framework scheduling workflow-automation webhook chat-interface developer-tools self-hosted mcp large-language-models developer-tools self-hosted cross-platform self-hosted cli go ai-agent-orchestration agent-runtime acp claude-code codex gemini-cli daemon agent-memory cron-scheduling multi-agent sqlite local-first beta ai-agents automation macos linux

10 sources

Member repositories

RepositoryRoleHealth v2
compozy/compozymain79

For agents

markdown · JSON · MCP: product_card(name="compozy/compozy")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem