Automatically apply safe quality fixes including formatting (Black, isort), linting (Ruff auto-fixes), and resolve formatter conflicts.
SkillScore68Maintained
Automatic
◆Coding pack
Last verified 9/2/2026
What it changes
Passed 11 of 16 test cases with the skill, and 4 of 16 without it.
Without skill23%
With skill68%
Checks it passed that the model alone missed
The response specifies passing --fix and --diff flags to Ruff.
The response recommends creating a Git commit prior to applying automated fixes.
The response specifies using the --format-only flag for selective layout fixing.
and 8 more in the full report
10 fixed5 passed either way7 still failing0 broke
Pass-rate lift
+45 pp
Tokens
+103%
Agent turns
0%
Response time
-53%
Model gemini-3.6-flashMeasured 9/2/2026v115 of 22 cases passed with skill · 16 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 10
Prompt
I want to trigger quality fixing on my project, but I only want to clean up import statements without touching code layout or variable syntax. What selective option flag should be passed?
What was checked, without and with the skill
Checked
Without skill
With skill
The response specifies using the --imports-only flag for selective import cleanup.
Missed
Passed
case-17 · gemini-3.6-flash · 9/2/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md582 linessingle file
Quality Fix Applier
⚠️ MANDATORY: Read Project Documentation First
BEFORE applying quality fixes, you MUST read and understand the following project documentation:
Core Project Documentation
README.md - Project overview, features, and getting started
AI_DOCS/project-context.md - Tech stack, architecture, development workflow
AI_DOCS/code-conventions.md - Code style, formatting, best practices
Verify the final deliverable meets all quality criteria before delivery. Use as the final validation step to ensure the output meets the user's quality standards across all 6 dimensions.