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

poco-ai/poco-claw

A more beautiful and easier-to-use alternative to OpenClaw. It features a nicer Web UI, built-in IM support, a sandboxed runtime and channel-based team collaboration. Under the hood, it is powered by a Claude Code–based agent. observed · 2026-08-28

github.com/poco-ai/poco-claw · homepage · Python · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

78/100

  • Activity 96
  • Release rhythm 89
  • Longevity 16
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: 5
  • age_days: 237
  • days_rel: 77
  • days_push: 27
  • n_releases_24m: 16

Full methodology

Adoption not part of the score

1351 stars · 127 forks observed · 2026-08-28

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

Poco is a self-hostable AI agent execution platform that serves as a more polished alternative to OpenClaw, powered by a Claude Code-based agent. It combines a Next.js web UI, FastAPI backend, sandboxed container execution, server/channel team collaboration, and IM integrations (DingTalk, Feishu, Telegram) in a Docker-deployable package.

Use cases

  • run a claude code agent through a web ui
  • self-host an openclaw alternative
  • collaborate with ai agents in channels and threads
  • execute agent tasks in a sandboxed container
  • control an ai agent from telegram or dingtalk
  • schedule background agent jobs in the cloud
  • let an agent browse the web and edit github repos
  • turn chat messages into trackable agent tasks

When to choose

  • you want a beautiful, self-hosted web UI for running Claude Code agents
  • you need team collaboration around agent runs with channels, DMs, and shared artifacts
  • you require sandboxed, isolated execution that protects the host machine
  • you want IM integration with DingTalk, Feishu, or Telegram
  • you need MCP, Skills, browser automation, and GitHub repo integration in one platform

When to avoid

  • you only need a lightweight CLI agent without a web UI or collaboration features
  • you cannot run Docker or self-host services
  • you want a fully managed cloud product and prefer not to operate infrastructure
  • you need an agent framework library to embed in your own application rather than a standalone platform

Facets

application · maturity active

agent-framework chatbot chat-interface mcp web-framework self-hosted large-language-models chatbots developer-tools self-hosted self-hosted python cross-platform claude-code openclaw-alternative sandboxed-execution im-integration team-collaboration channel-based dingtalk feishu telegram nextjs fastapi task-management artifacts-view browser-automation sandboxed-runtime ai-agents automation docker web-server

2 sources

Member repositories

RepositoryRoleHealth v2
poco-ai/poco-clawmain78

For agents

markdown · JSON · MCP: product_card(name="poco-ai/poco-claw")

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