Safe Autonomous Code Fixing

What you'll learn

• Design an automated bug-fixing pipeline that reproduces, localises, patches, verifies and hands off to review.
• Reproduce bugs with failing tests and localise them with stack traces and git bisect.
• Constrain patches to minimal diffs and verify them in isolated environments with flaky-test and mutation checks.
• Block unsafe patches: test tampering, protected paths, leaked secrets and over-broad permissions.
• Ship automated fixes through human review with clear evidence, and revert safely when needed.
• Measure an auto-fix system with merge, revert and review-time metrics.

Syllabus

Why Autonomous Fixing Needs Safety

1. Where Automated Fixes Help
2. What Can Go Wrong
3. The Safe Fix Pipeline

Reproduce and Localise

1. Reproducing With a Failing Test
2. Reading Stack Traces and Logs
3. Finding the Breaking Commit With git bisect

Generating Minimal Patches

1. Giving the Fixer the Right Context
2. Minimal Diffs and Size Limits
3. Multiple Candidates and Selection

Verifying Patches

1. Isolated Verification of Each Candidate
2. Detecting Flaky Tests
3. Are the Tests Strong Enough? Mutation Checks
4. Static Checks, Types and Builds

Guardrails on Patches

1. Detecting Test Tampering
2. Protected Paths and File Allowlists
3. Secret Scanning on Every Patch

Sandboxes and Permissions

1. Sandboxed Execution
2. Least-Privilege Tokens and Branch Protection
3. Untrusted Instructions in Issues and Logs

Review, Merge and Revert

1. Pull Requests With Evidence
2. Review and Merge Policy
3. Reverting Safely

Dependency Fixes, Metrics and a Checklist

1. Automated Dependency and Security Fixes
2. Measuring an Auto-Fix System
3. A Safe Auto-Fix Checklist