openclaw-github-repo-commander – skill Lập trình cho agent AI

OpenClaw GitHub Repo Commander

Overview

A structured 7-stage super workflow for comprehensive GitHub repository management. This skill automates repository auditing, cleanup, competitor benchmarking, and optimization — turning a messy repo into a clean, well-documented, production-ready project.

When to Use This Skill

• Use when you need to audit a repository for secrets, junk files, or low-quality content
• Use when the user says "clean up my repo", "optimize my GitHub project", or "audit this library"
• Use when reviewing or creating pull requests with structured analysis
• Use when comparing your project against competitors on GitHub
• Use when running /super-workflow or /openclaw-github-repo-commander on a repo URL

How It Works

Stage 1: Intake

Clone the target repository, define success criteria, and establish baseline metrics.

Stage 2: Execution

Run bash scripts/repo-audit.sh <repo-path> from this skill directory — automated read-only checks for:

• Hardcoded secrets (ghp_, sk-, AKIA, etc.)
• Tracked node_modules/ or build artifacts
• Empty directories
• Large files (>1MB)
• Missing .gitignore coverage
• Broken internal README links

Stage 3: Reflection

Deep manual review beyond automation: content quality, documentation consistency, structural issues, version mismatches.

Stage 4: Competitor Analysis

Search GitHub for similar repositories. Compare documentation standards, feature coverage, star counts, and community adoption.

Stage 5: Synthesis

Consolidate all findings into a prioritized action plan (P0 critical / P1 important / P2 nice-to-have).

Stage 6: Iteration

Execute the plan: delete low-value files, fix security issues, upgrade documentation, add CI workflows, update changelogs.

Stage 7: Validation

Re-run applicable checks and report exact coverage/warnings. The bundled helper has six broad checks, excludes workflows and distinguishes neither all tracked/local files nor every secret format; a clean result is not security certification. Push or PR only through the actual repository workflow and user-authorized scope.

Examples

Example 1: Full Repo Audit

/openclaw-github-repo-commander https://github.com/owner/my-repo

Runs all 7 stages and produces a detailed before/after report.

Example 2: Quick Cleanup

Clean up my GitHub repo — remove junk files, fix secrets, add .gitignore

Example 3: Competitor Benchmarking

Compare my skill repo with the top 5 similar repos on GitHub

Best Practices

• ✅ Always run Stage 7 validation before pushing
• ✅ Use semantic commit messages: chore:, fix:, docs:
• ✅ Check the pr_todo.json file for pending reviewer requests
• ❌ Don't skip Stage 4 — competitor analysis reveals blind spots
• ❌ Don't commit node_modules/ or .env files

Security & Safety Notes

The helper distinguishes a clean no-match from scanner failure and reports failure as unavailable. It still uses limited heuristic patterns and exclusions; do not execute it on untrusted targets as a review shortcut or treat its output as complete secret coverage.

• The audit script scans for common secret patterns but excludes .github/workflows/ to avoid false positives
• The bundled script is read-only: it reports findings and never deletes, rewrites, stages, commits, or pushes files
• All gh CLI operations use the user's existing authentication — no credentials are stored by this skill
• The skill never modifies files without explicit user confirmation in Stage 6

Source Attribution

Originally contributed by [@wd041216-bit](https://github.com/wd041216-bit) in [PR #340](https://github.com/sickn33/agentic-awesome-skills/pull/340). No standalone upstream repository is currently available for this skill.

License: MIT | Version: 4.0.0

Limitations

• Use this skill only when the task clearly matches the scope described above.
• Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
• Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.