One Sentence From a Human, Four Agents Fixed the Bug

Description and use

Four AI agents in a shared workspace quickly fixed a download bug, caught real issues, and deployed the fix without manual intervention.

Problem

A download method silently failed under certain conditions, losing the filename, requiring a bug fix that normally would take hours of human coordination.

Outcome

The four‑agent pipeline fixed the bug in minutes, caught a real issue that would have caused silent data corruption, and deployed the fix without manual intervention.

Workflow

• Human raises the issue with a single sentence: "Confirm whether this bug is real. If so, fix it per the process."
• PM agent reads the code, identifies root cause, plans fix, assigns task to Developer.
• Developer agent writes code, creates PR, and submits.
• Reviewer agent reviews PR, finds two issues (filename loss and error page save), requests changes.
• Developer fixes issues, updates PR.
• Reviewer re‑reviews and approves.
• Deployer agent packages and deploys to test environment.
• PM generates acceptance checklist and asks human to click‑test.
• Human verifies and accepts the fix.

Evidence

The PM agent didn't wait for someone to explain the bug. It read the code itself, found the root cause.

Source

The Reviewer found two issues. One of them: if the file download fails, the existing code would save an error page as if it were the actual file.

Source

The whole point is: the human's role shifts from doing the work to defining the outcome and accepting it.

Source