Open-Source AI Agent Framework on GitHub Actions
I run an agent that writes me a morning brief before I'm out of bed. It tracks the repos I care about and stays quiet on slow days. It pays for the web searches it needs out of its own wallet. It does all of this on a schedule, in GitHub Actions, on my own copy of the aeon public repo. No server. No always-on laptop. No SaaS dashboard charging me by the seat. This post kicked off a series on why…
A new open-source AI agent framework called Aeon has been launched on GitHub Actions. The framework allows users to create autonomous agents that run on GitHub Actions runners, using their own GitHub repositories. Users can sign in with GitHub at Aeon Connect, connect a model, and select skills to create an agent that operates on a schedule, without requiring a server or always-on laptop.
The agent can perform various tasks such as tracking repositories, monitoring web searches, and generating reports, all without the need for a dashboard or paying by the seat. Aeon offers a range of skills grouped into six packs, which are single Markdown files containing prompts and instructions. Users can edit existing skills, create new ones, and even chain skills together.
The framework is designed to be self-healing, with built-in mechanisms to handle failed runs and monitor the overall health of the agent. One of the key differentiators of Aeon is its ability to not only fetch and summarize information but also perform actions in the real world, such as making payments and accessing premium feeds.
The framework integrates with Base MCP and the x402 protocol, allowing agents to spend and earn money through the Base Account wallet. Skills in Aeon are Markdown files, making them easy to edit, compose, and improve over time. The framework also provides a dashboard for onboarding and monitoring the agent's activities. Users can start using Aeon by creating an account on Aeon Connect and connecting their GitHub account to create their first agent.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
