Install any skill in seconds. Free to start, no credit card required.
Get Started Free →7-stage super workflow for GitHub repo audit, cleanup, PR review, and competitor analysis
.claude/skills/lingxling-openclaw-github-repo-commander/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -40% | 0% |
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.
/super-workflow or /openclaw-github-repo-commander on a repo URLClone the target repository, define success criteria, and establish baseline metrics.
Run scripts/repo-audit.sh <repo-path> from this skill directory — automated read-only checks for:
ghp_, sk-, AKIA, etc.)node_modules/ or build artifacts.gitignore coverageDeep manual review beyond automation: content quality, documentation consistency, structural issues, version mismatches.
Search GitHub for similar repositories. Compare documentation standards, feature coverage, star counts, and community adoption.
Consolidate all findings into a prioritized action plan (P0 critical / P1 important / P2 nice-to-have).
Execute the plan: delete low-value files, fix security issues, upgrade documentation, add CI workflows, update changelogs.
Re-run the audit script (target: 7/7 PASS), verify all changes, push to GitHub, and deliver a full report.
/openclaw-github-repo-commander https://github.com/owner/my-repoRuns all 7 stages and produces a detailed before/after report.
Clean up my GitHub repo — remove junk files, fix secrets, add .gitignoreCompare my skill repo with the top 5 similar repos on GitHubchore:, fix:, docs:pr_todo.json file for pending reviewer requestsnode_modules/ or .env files.github/workflows/ to avoid false positivesgh CLI operations use the user's existing authentication — no credentials are stored by this skillOriginally contributed by @wd041216-bit in PR #340. No standalone upstream repository is currently available for this skill.
License: MIT | Version: 4.0.0
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,696 | 29,387 | +42% | 1 | 1 | 0% | 2,900 | 3,902 | +35% | 0 | 0 | — |
case-02 | fail→fail | 27,116 | 21,600 | -20% | 1 | 1 | 0% | 1,146 | 4,059 | +254% | 0 | 0 | — |
case-03 | fail→fail | 20,828 | 42,644 | +105% | 1 | 1 | 0% | 3,423 | 7,062 | +106% | 0 | 0 | — |
case-04 | pass→pass | 15,216 | 18,934 | +24% | 1 | 1 | 0% | 3,220 | 4,197 | +30% | 0 | 0 | — |
case-05 | pass→pass | 211,734 | 19,154 | -91% | 1 | 1 | 0% | 4,032 | 3,760 | -7% | 0 | 0 | — |
case-06 | pass→pass | 18,497 | 20,813 | +13% | 1 | 1 | 0% | 2,869 | 3,762 | +31% | 0 | 0 | — |
case-07 | fail→pass | 11,869 | 4,978 | -58% | 1 | 1 | 0% | 1,508 | 1,605 | +6% | 0 | 0 | — |
case-08 | fail→pass | 15,261 | 8,562 | -44% | 1 | 1 | 0% | 2,473 | 1,659 | -33% | 0 | 0 | — |
case-09 | pass→pass | 12,707 | 8,352 | -34% | 1 | 1 | 0% | 2,042 | 1,839 | -10% | 0 | 0 | — |
case-10 | pass→pass | 69,270 | 10,363 | -85% | 1 | 1 | 0% | 2,416 | 2,552 | +6% | 0 | 0 | — |
case-11 | fail→pass | 17,504 | 4,259 | -76% | 1 | 1 | 0% | 1,204 | 1,467 | +22% | 0 | 0 | — |
case-12 | fail→pass | 11,319 | 4,332 | -62% | 1 | 1 | 0% | 1,461 | 1,521 | +4% | 0 | 0 | — |
case-13 | fail→pass | 12,102 | 3,115 | -74% | 1 | 1 | 0% | 2,065 | 1,249 | -40% | 0 | 0 | — |
case-14 | fail→pass | 9,230 | 3,038 | -67% | 1 | 1 | 0% | 1,444 | 1,265 | -12% | 0 | 0 | — |
case-15 | pass→pass | 13,263 | 8,026 | -39% | 1 | 1 | 0% | 2,584 | 1,539 | -40% | 0 | 0 | — |
case-16 | fail→pass | 21,203 | 14,459 | -32% | 1 | 1 | 0% | 3,189 | 3,224 | +1% | 0 | 0 | — |
case-17 | fail→pass | 17,263 | 13,149 | -24% | 1 | 1 | 0% | 2,522 | 2,148 | -15% | 0 | 0 | — |
case-18 | pass→pass | 17,428 | 10,360 | -41% | 1 | 1 | 0% | 2,128 | 2,530 | +19% | 0 | 0 | — |
case-19 | fail→pass | 10,963 | 4,044 | -63% | 1 | 1 | 0% | 1,448 | 1,596 | +10% | 0 | 0 | — |
case-20 | fail→pass | 10,724 | 120,375 | +1022% | 1 | 1 | 0% | 1,723 | 1,871 | +9% | 0 | 0 | — |
case-21 | pass→pass | 14,396 | 2,631 | -82% | 1 | 1 | 0% | 1,919 | 1,196 | -38% | 0 | 0 | — |
case-22 | fail→pass | 19,350 | 8,478 | -56% | 1 | 1 | 0% | 2,672 | 2,148 | -20% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.