Repository Manager

Automate all claude-skills-worth-using repository operations.

Activation Triggers

Status & Health:

"Check claude-skills-worth-using status"
"Update claude-skills-worth-using"
"Show repository health"

Adding Skills:

"Add [skill-name] from [source]"
"Find and add [category] skills"
"Integrate skills from [repo-url]"

Documentation:

"Update skills repo README"
"Update README to version [X.X.X]"
"Refresh documentation"

Releases:

"Release version [X.X.X]"
"Create release for skills repo"
"Tag version [X.X.X]"

Maintenance:

"Clean up skills repository"
"Check for skill updates"
"Validate repository structure"

Core Workflows

1. Repository Status Check (Always Start Here)

When user mentions the repository, first check current state:

1. Scan skills directory:
   github:get_file_contents(owner="Organized-AI", repo="clalude-skills-worth-using", path="skills")

2. Count and categorize skills:
   - ⏰ Productivity & Planning
   - 💻 Development Tools
   - 💼 Business & M&A
   - 🔍 Content & Marketing
   - 🧠 Knowledge & Patterns

3. Check README version:
   github:get_file_contents(owner="Organized-AI", repo="clalude-skills-worth-using", path="README.md")
   Extract: Version number, last update date

4. Check for open PRs:
   github:list_pull_requests(owner="Organized-AI", repo="clalude-skills-worth-using", state="open")

5. Report status:
   "Repository has X skills across Y categories (version Z.Z.Z)"
   "Health: [Excellent/Good/Needs Attention]"
   "Recommendations: [list any needed actions]"

2. Adding Skills from External Sources

When user wants to add skills:

Phase 1: Validate Source
1. Verify license (MUST be MIT or compatible)
2. Confirm skills to add with user
3. Check for duplicates in existing repo

Phase 2: Prepare Files
1. Download complete skill structure:
   - SKILL.md (2-5KB, core file)
   - README.md (if exists)
   - /references directory
   - /examples directory
   - /assets or /scripts

2. Clean files:
   - Remove .DS_Store, node_modules, package.json, index.js
   - Verify SKILL.md has proper structure
   - Check size (should be 2-5KB, modular)

Phase 3: Push to Repository
1. Create branch: feature/add-[skill-names]
2. Use github:push_files to upload all at once
3. Create/update attribution file if external source
4. Create comprehensive PR (see references/pr-templates.md)

Phase 4: Quality Check
- Validate structure (2-5KB, modular)
- Check activation phrases
- Verify documentation complete
- Test no file conflicts

3. Updating Main README

When skills added or repository changes:

1. Scan all skills in /skills directory

2. Categorize each skill:
   Read SKILL.md to determine category based on content

3. Generate skill sections:
   For each skill:
   - Icon + name
   - One-line description
   - Feature bullets
   - Activation examples
   - Tool requirements
   - Documentation link

4. Update comparison matrix:
   Table with: Skill, Category, Automation, Setup, Best For

5. Update repository structure:
   Show correct skill count

6. Increment version:
   - 1-2 skills added → Patch (x.x.X)
   - 3-5 skills added → Minor (x.X.0)
   - 6+ skills added → Major (X.0.0)

7. Create branch and PR:
   Branch: feature/update-readme-[version]
   Use comprehensive PR template

See references/workflows.md for complete workflow details.

4. Version Management

When releasing:

1. Determine version number:
   See references/version-logic.md for decision tree

2. Update version references:
   - README.md: "Version: X.X.X"
   - README.md: "Last Updated: [date]"

3. Create release:
   - Branch: release/vX.X.X
   - Tag: git tag -a vX.X.X -m "Summary"
   - GitHub release with changelog

4. Offer announcement content

5. Quality Assurance

Before any PR or release:

Run validation checklist from references/quality-checklist.md

Key checks:

✅ All skills 2-5KB (modular)
✅ Documentation 100% coverage
✅ Attribution files present (for external skills)
✅ No broken links
✅ Consistent naming
✅ Version current

6. Repository Maintenance

Periodic tasks:

Weekly/Monthly:
- Check external skills for updates
- Verify all links work
- Clean unnecessary files
- Update dependencies

Cleanup:
- Remove .DS_Store files
- Verify directory consistency
- Update attribution if needed

See references/maintenance-schedule.md for complete schedule.

Tool Requirements

Required GitHub MCP:

github:get_file_contents
github:create_branch
github:create_or_update_file
github:push_files
github:create_pull_request
github:list_pull_requests

Recommended:

Memory system (track state)
Sequential thinking (complex workflows)

Decision Logic

When to Update README

Skill added → Yes
Skill removed → Yes
Category changed → Yes
Version released → Yes

Version Number Selection

1-2 skills? → Patch (x.x.X)
3-5 skills? → Minor (x.X.0)
6+ skills? → Major (X.0.0)
Breaking changes? → Major (X.0.0)
Docs only? → Patch (x.x.X)

Branch Naming

Adding skills? → feature/add-[names]
Updating docs? → feature/update-readme-[version]
Releasing? → release/v[version]
Bug fix? → fix/[description]

Best Practices

Always verify license first - No exceptions
Check status before actions - Know current state
Use atomic commits - One logical change per PR
Update attribution immediately - Don't forget credits
Validate before pushing - Check structure and links
Quality over quantity - 10 great skills > 50 mediocre

Expected Outcomes

Time Savings:

Add skills: 30 min → 3 min (90%)
Update README: 45 min → 2 min (96%)
Release version: 15 min → 2 min (87%)
Status check: 10 min → 30 sec (95%)

Quality Guarantees:

100% structure compliance
100% documentation coverage
100% attribution accuracy
100% naming consistency

Error Handling

Common Issues:

Skill exists → Check for duplicates first
License incompatible → Don't add, inform user
README out of sync → Run README update workflow
Broken links → Validate and fix
File conflicts → Choose newer/better version

Integration

Works with:

Memory - Track repository state
Weekly Planner - Schedule maintenance
Tech Debt Analyzer - Assess repo health
All GitHub Tools - Full GitHub integration

Example Commands

"What's the current state of claude-skills-worth-using?"
→ Runs status check, provides health report

"Add codebase-documenter skill from ailabs-393"
→ Downloads, cleans, attributes, pushes, creates PR

"Update README to reflect all current skills"
→ Scans, categorizes, generates sections, creates PR

"Create version 3.1.0 release"
→ Determines version, tags, creates release, offers announcement

"Check all skills for updates"
→ Scans external sources, reports available updates

Success Metrics

Track:

Time saved per operation (target: 85-95%)
Documentation coverage (target: 100%)
Structure compliance (target: 100%)
Attribution accuracy (target: 100%)

For detailed workflows, templates, and examples:

See references/workflows.md - Complete step-by-step procedures
See references/pr-templates.md - PR and commit templates
See references/decision-trees.md - All decision logic
See references/quality-checklist.md - Complete validation checklist

repo-manager