af-fix-analyze

To automatically fix lint and analysis issues iteratively, you need a tool that reads your analysis document, applies targeted code changes to resolve the highest-severity issue, and verifies the fix with build and lint checks before committing. This Skill automates that exact workflow by modifying code, running verifications, updating the analysis file, and creating a commit prefixed with the issue ID.

The best way to apply safe bug fixes from an analysis document is to process them iteratively. This Skill selects the highest-severity unfixed issue, attempts a minimal code change, runs build and lint checks to verify safety, documents the outcome with reasons, and updates the analysis record with issue metadata.

Yes, this automated code fix approach supports security and technical debt analysis. It is designed to handle multiple analysis types including bug, debt, doc, security, and style, reading the specific issue lists from your analysis files and applying targeted fixes during maintenance workflows.

Iterative issue resolution with autofix works by reading the analysis file, selecting the top prioritized unfixed issue, applying a minimal code modification, running build and lint verifications, and creating a commit. The analysis document is then updated to reflect whether the issue was fixed or skipped.

If a build or lint check fails after applying an automated code fix, the issue is documented as skipped with specific reasons in the analysis file. The workflow ensures only verified changes are committed, preventing broken code from entering the repository while maintaining a record of the attempt.

Yes, you need an existing analysis file to start fixing repository issues automatically. The Skill operates by reading prioritized issue lists from the agentic/analysis directory, requiring a pre-existing document to identify the highest-severity unfixed issue and apply the corresponding code changes.