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

clacky-ai/openclacky

The most Token-efficient open-source AI Agent observed · 2026-09-03

github.com/clacky-ai/openclacky · homepage · Ruby · MIT (permissive) observed · 2026-09-03

Health v2 · maintenance only

79/100

  • Activity 100
  • Release rhythm 87
  • Longevity 17
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: 0
  • age_days: 246
  • days_rel: 6
  • days_push: 0
  • n_releases_24m: 126

Full methodology

Adoption not part of the score

1184 stars · 105 forks observed · 2026-09-03

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

OpenClacky is an open-source (MIT) AI agent platform written in Ruby that runs locally, offering a CLI and Web UI for completing tasks via Skills (capability packages), file/command/browser access, and long-term memory. It emphasizes token efficiency through a minimal 16-tool set, near-100% prompt cache hit rates, and BYOK support for any OpenAI-compatible model.

Use cases

  • run an AI agent locally that can read and write files and execute commands
  • reduce LLM token costs when using an open-source coding or task agent
  • use Claude Code-style agent with my own API key (BYOK)
  • automate recurring tasks like weekly reports or news digests with scheduled agent runs
  • integrate an AI agent into Feishu, WeCom, WeChat, Discord, or Telegram
  • create and share reusable AI workflow skills or publish them to a marketplace
  • control a browser to automate web tasks with an AI agent

When to choose

  • you want a Claude Code-class open-source agent with lower token spend
  • you need BYOK model freedom across OpenAI, Anthropic, Google, and other providers
  • you want local execution with file, command, and browser access plus a Web UI and CLI
  • you want a skill ecosystem with self-evolving, shareable capability packages

When to avoid

  • you need a fully managed cloud agent service without running anything locally
  • you require Anthropic-only subscription-style billing instead of BYOK
  • you need a mature multi-agent orchestration framework rather than a single efficient agent
  • your stack requires a Python or JavaScript agent SDK rather than a Ruby-based tool

Facets

application · maturity active

agent-framework llm-inference cli chatbot mcp workflow-automation browser-extension large-language-models developer-tools windows cross-platform cli ruby ai-agent token-efficiency byok skills skill-marketplace im-integration browser-automation local-agent prompt-caching open-source-agent ai-agents automation command-line macos linux web-server

7 sources

Member repositories

RepositoryRoleHealth v2
clacky-ai/openclackymain79

For agents

markdown · JSON · MCP: product_card(name="clacky-ai/openclacky")

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