Last Wednesday (Sep 10), I had the chance to attend a hackathon at Cursor’s offices - huge thanks to the Cursor team for hosting such a great event! 🙌 The focus was on Cursor CLI and their new Background Agents API.
When I started brainstorming ideas, I came across Eric’s post about running various rules during CI/CD to automate checks and actions. It got me thinking — setting up those rules often involves a lot of repetitive work: crafting similar prompt setups and then defining the actions themselves.
That’s when the idea for AutoAgent Action was born. 🚀
What is AutoAgent Action?
AutoAgent is a GitHub Action that makes it easy to run checks using Cursor CLI (or other AI CLIs like Claude Code, Codex CLI, Gemini CLI, or Amp). It ships with a collection of preset rules that are immediately useful to many developers, and it’s completely customizable.
With AutoAgent, you can:
- ✅ Enable built-in rules with minimal setup
- 🛠 Add your own custom rules directly in your workflow
- 📂 Point to a folder in your repo that contains your own rule set
How It Works
When AutoAgent runs, it:
- Executes the selected rules against your pull request.
- Posts a detailed comment on the PR with issues, insights, and suggestions.
- (Optional) Launches a Cursor background agent to automatically work on the PR and fix issues for you.
You control how hands-on or hands-off you want to be — from passive checks to fully automated fixes.
Why I Built It
While exploring Eric’s post of rule-based CI/CD checks, I realized most setups require a lot of duplicate work:
- Writing nearly identical prompt boilerplate for each rule
- Wiring each rule to a separate action or script
- Maintaining everything across multiple repos
AutoAgent streamlines this entire process by providing a unified, flexible way to manage rules and actions.
Try It Out
Check out the project here: github.com/erans/autoagent-action
I’d love your feedback! If you find AutoAgent useful, please consider giving it a ⭐ — it really helps support the project.
Thanks again to the Cursor team for hosting such an inspiring hackathon and to everyone who shared ideas that helped bring AutoAgent to life.