#589 · Primary category: AI Agents & Automation

aeon

agent-framework agents ai ai-agents anthropic automation autonomous-agents claude claude-code devops github-actions harness llm mcp

The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever.

Project last updated:08/29/26

GitHub Stars

705

Forks

252

Contributors

69

License

MIT

Why we included this project

If you want an agent that keeps working while you step away, Aeon is built around that exact trade-off: it runs unattended on GitHub Actions, so a scheduled skill can ship a feature, run deep research, or scan for vulnerabilities without anyone approving each step. Skills are plain Markdown files with a prompt and some frontmatter, which keeps them easy to read, version, and share across a team. The same skill definition runs on any of nine agent CLIs, from Claude Code and Codex to Grok and Cursor, behind a single harness contract, so you are not locked into one vendor's tooling. It also scores every run and repairs its own broken skills, which matters if you plan to leave long-running automations in place. Teams that already live in GitHub and want scheduled, self-healing agent work with guardrails will find the setup path unusually direct.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category