Use an agent to patch a stubborn app bug
They pointed a coding agent at two annoying software behaviors and had it debug one and add a self-improvement feature to the other.
- What they did
- A developer aimed Codex and ChatGPT at software that was bothering them. In one case, the agent tracked down a Hyper-V vmconnect fullscreen problem after login and built a hook-based fix that did not alter Microsoft’s files. In another, they had Codex work on the Codex Electron app itself to add an MCP-driven setup with dynamic JavaScript userscripts.
- How it went
- Both changes reportedly worked, though the agent sometimes needed several rounds of nudging to improve the result. It still struggled to write documentation and reports that people could read easily.
- Tips
- This works best when you give the agent a narrow bug or missing feature and ask it to test the fix before any changes land. Make sure it pauses for your approval before anything that can’t be easily undone.
Try it yourself with Codex
I have a bug or missing feature in [app or tool name]: [describe the problem]. Inspect the code or install location at [path or repo] and find the smallest safe change or hook that would fix it. First, show me the likely cause, the patch plan, and how I can test it; then wait for my approval before editing, sending, deleting, or submitting anything. If your first attempt is off, revise it and explain what changed. When it works, include the diff and a quick rollback plan.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Use spare AI quota to draft reviewable PRs
A founder pointed an agent at goals and task lists so it could turn leftover Claude and Codex quota into pull requests for review.
Monitor account caps and move chats across tools
teivz built a local app that watches Codex and Claude Code account usage and shifts a chat to another account when one reaches its cap.
Use an AI spec to brief a coding agent
Work out a feature with an AI assistant, then pass the spec to a coding agent and verify the result by testing.
Create and launch a BaZi reading agent
A developer used coding agents to build a BaZi reader that calculates charts, writes interpretations, and handles language detection and USDT payments.
Investigate why a service is getting killed
Run a read-only SSH check to find the cause of a process death and rank fixes by risk.
Flag coding conflicts before agents edit
They set up a local layer that makes coding agents announce planned changes so overlapping architecture work gets caught early.