4meta5

repo-hygiene

0
0
# Install this skill:
npx skills add 4meta5/skills --skill "repo-hygiene"

Install specific skill from multi-skill repository

# Description

|

# SKILL.md


name: repo-hygiene
description: |
Detect and clean AI-generated slop from skills directories. Use when: (1) test-skill-*
directories accumulate from testing, (2) CLAUDE.md has stale skill references,
(3) placeholder content exists in skills. Provides scan, clean, and CLAUDE.md sync.
category: development
user-invocable: true


Repo Hygiene

Detect and clean AI-generated slop (test skills, placeholder content) from your project.

Quick Start

# Scan for slop
skills hygiene scan

# Scan including package subdirectories
skills hygiene scan -r

# Preview what would be deleted
skills hygiene clean --dry-run

# Actually delete slop
skills hygiene clean --confirm

What is Slop?

Slop is auto-generated or placeholder content that shouldn't be committed:

Pattern Example Action
test-skill-* test-skill-1234567890 Delete
Timestamped my-skill-1706625000000 Review
_temp_* _temp_claude-svelte5-skill Review (may need rename)
Placeholder "NEW content with improvements!" Delete

Detection Patterns

test-skill-* (Auto-Delete)

Skills matching /^test-skill-\d+$/ are auto-generated test data.
These are always safe to delete.

Timestamped Names (Review)

Skills ending with 13-digit timestamps (-1706625000000) may be
auto-generated. Review before deleting.

temp Prefix (Review)

Skills starting with _temp_ were likely created as temporary work.
They may contain valuable content that needs proper naming.

Recommended action: Rename to remove prefix if content is valuable,
or delete if it's truly temporary.

Placeholder Content (Auto-Delete)

Skills containing these patterns are incomplete:
- # Test Skill (exact match)
- NEW content with improvements!

Slop Locations

The scan checks these locations:

  1. .claude/skills/ - Root project skills
  2. packages/*/\.claude/skills/ - Package-level skills (with -r)

CLAUDE.md Cleanup

The scan also checks CLAUDE.md for:

  • Stale references: Skills listed but not installed
  • Duplicate references: Same skill listed multiple times

Run skills claudemd sync to fix CLAUDE.md references.

Workflow

Daily Cleanup

After testing or development:

skills hygiene scan
skills hygiene clean --confirm
skills claudemd sync

Pre-Commit Check

Add to your workflow:

# Check for slop before committing
skills hygiene scan
if [ $? -ne 0 ]; then
  echo "Slop detected! Run: skills hygiene clean --confirm"
  exit 1
fi

Decision Tree

Found test-skill-*?
├─ YES → Delete (safe)
└─ NO
   ├─ Found _temp_* with good content?
   │  └─ Rename to proper name, delete _temp_ version
   ├─ Found placeholder content?
   │  └─ Delete or rewrite
   └─ Found stale CLAUDE.md refs?
      └─ Run: skills claudemd sync

Safety

  • --dry-run shows what would happen without changes
  • --confirm is required for actual deletion
  • _temp_* skills are never auto-deleted (review required)
  • Timestamped skills are flagged for review, not auto-deleted
  • skills validate - Check skill quality
  • skills claudemd sync - Sync CLAUDE.md with installed skills
  • skills list - Show installed skills

Skill Chaining

Works with:
- skill-maker: Create clean skills to replace slop
- claudeception: Extract learnings before deleting temp skills
- workflow-orchestrator: Include hygiene in project workflow

Terminal Chain (Always Run Last)

This skill runs after testing workflows complete:

After Skill Cleanup Needed
tdd test-skill-* directories
suggest-tests Analysis artifacts
unit-test-workflow Generated test stubs
property-based-testing Fast-check artifacts
doc-maintenance Stale references
gitignore-hygiene Final verification

When any testing skill completes, run:
- skills hygiene scan
- skills hygiene clean --confirm (if slop detected)

Testing Pipeline Position

repo-hygiene is the terminal step in the testing pipeline:

tdd → suggest-tests → unit-test-workflow → property-based-testing → repo-hygiene
                                                                         ↑
                                                                    (TERMINAL)

All testing workflows should end with repo-hygiene to ensure clean state.

Feature Completion Flow

In feature completion contexts, repo-hygiene chains to doc-maintenance:

feature done → dogfood-skills → repo-hygiene → doc-maintenance

After cleanup, doc-maintenance updates PLAN.md to mark tasks complete.

# Supported AI Coding Agents

This skill is compatible with the SKILL.md standard and works with all major AI coding agents:

Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.