Fixes a single file by applying minimal, safe edits while running linters and tests to verify the change.

Install

mkdir -p .claude/skills/fix-file && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/18441" && unzip -o skill.zip -d .claude/skills/fix-file && rm skill.zip

Installs to .claude/skills/fix-file

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Use when you need to fix one file end-to-end: lint, test impact, and apply minimal safe changes for this project.
113 chars · catalog description✓ has a “when” trigger
Intermediate

Key capabilities

  • Identify failing behavior and minimal edit scope
  • Apply focused changes without unrelated refactors
  • Run targeted tests first, then broader checks
  • Report exactly what changed and why
  • Enforce project guardrails like no suppression patterns
  • Preserve atomic-save semantics

How it works

The skill identifies failing behavior, applies minimal changes to a target file, and validates these changes through targeted tests and broader checks.

Inputs & outputs

You give it
A target file path and a description of failing behavior
You get back
A validated, minimally changed target file and a report of changes

When to use fix-file

  • Fixing a specific linting error in a single file
  • Correcting a isolated bug detected by a unit test
  • Applying formatting changes that pass CI checks
  • Executing targeted tests after a minor logic patch

About this skill


name: fix-file user-invocable: true description: Use when you need a targeted single-file fix workflow for one file end-to-end, including lint checks, impacted tests, and minimal safe edits.


Fix File Skill

Goal

Apply the smallest safe fix to a target file and validate it.

Workflow

  1. Identify the failing behavior and the minimal edit scope.
  2. Apply focused changes without unrelated refactors.
  3. Run targeted tests first, then broader checks if needed.
  4. Report exactly what changed and why.

Commands

poetry run python tests/tools/check_no_suppressions.py
poetry run ruff check <target_file_path>
poetry run ruff format --check <target_file_path>
poetry run flake8 <target_file_path>
poetry run pytest <target_test_path>
.\run_local_pipeline.ps1

Project-Specific Guardrails

  • Keep orchestration in auto_subtitle.py and logic in modules/.
  • Do not weaken shutdown and process cleanup behavior.
  • Preserve resume and atomic-save semantics.
  • Keep complexity below 10.
  • Never add suppression patterns (noqa, type: ignore, warning-ignore filters).

When not to use it

  • When changes are not minimal
  • When changes involve unrelated refactors
  • When adding suppression patterns

Limitations

  • The skill does not allow adding suppression patterns
  • The skill requires preserving atomic-save semantics
  • The skill enforces keeping complexity below 10

How it compares

This workflow focuses on minimal, validated changes to a single file, unlike a manual process that might introduce broader, unvalidated modifications.

Compared to similar skills

fix-file side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
fix-file (this skill)010dNo flagsIntermediate
python-testing-patterns772moReviewIntermediate
clojure-write162moNo flagsIntermediate
python-playground24moReviewBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry