Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Scaffold or standardize a production-ready repository structure with specs, source layout, tests, CI, agent context, config examples, release notes, and operational docs. Use when starting a new repo, turning a prototype into a maintainable project, adding missing repository foundations, or creating a repo skeleton before implementation. For AGENTS.md-only context scaffolding, use agentsmd-scaffold.
.claude/skills/majiayu000-repo-scaffold/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 165% | 0% |
| case-09 | ✓→✗ | ▼ Worse | -50% | 0% |
Use this skill to create the repo foundation that lets product, architecture, implementation, verification, and operations stay connected. It is broader than agentsmd-scaffold; it may include AGENTS.md, but also specs, tests, CI, config, release, and runbook structure.
Before adding files, inspect the repo for existing equivalents:
bashrg --files -g 'AGENTS.md' -g 'CONTRIBUTING.md' -g 'README*' -g 'pyproject.toml' -g 'package.json' -g 'go.mod' -g 'Cargo.toml' -g '.github/workflows/*' -g 'docs/**' -g 'specs/**' -g 'tests/**'
Reuse existing conventions. Do not create parallel docs/, spec/, planning/, or test/ trees when the repo already has a standard location.
Add only the layers needed for the repo:
| Layer | Typical Files | |---|---| | Product and specs | specs/PRODUCT.md, specs/TECH.md, docs/adr/ | | Agent context | AGENTS.md, scoped AGENTS.md, local skill notes | | Source layout | language-specific src/, pkg/, cmd/, app/, lib/ | | Tests | tests/, fixtures, golden snapshots, e2e harness | | CI | .github/workflows/ci.yml, lint/typecheck/test jobs | | Config | .env.example, config schema, secret inventory | | Release | CHANGELOG.md, release checklist, versioning notes | | Operations | docs/runbooks/, SLO and incident templates |
For planning:
textexisting_foundation: missing_layers: proposed_tree: files_to_create_or_update: verification_commands:
For implementation, finish by running the repo's validation command and, when available, the scaffold-specific lint or generated-file check.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 17,071 | 6,091 | -64% | 1 | 1 | 0% | 2,687 | 877 | -67% | 0 | 0 | — |
case-02 | fail→fail | 21,935 | 3,750 | -83% | 1 | 1 | 0% | 3,975 | 958 | -76% | 0 | 0 | — |
case-03 | fail→fail | 4,035 | 5,780 | +43% | 1 | 1 | 0% | 204 | 922 | +352% | 0 | 0 | — |
case-04 | fail→fail | 10,066 | 4,990 | -50% | 1 | 1 | 0% | 1,781 | 1,391 | -22% | 0 | 0 | — |
case-05 | fail→pass | 14,300 | 8,166 | -43% | 1 | 1 | 0% | 2,215 | 1,353 | -39% | 0 | 0 | — |
case-06 | pass→pass | 13,112 | 8,425 | -36% | 1 | 1 | 0% | 2,124 | 1,901 | -10% | 0 | 0 | — |
case-07 | pass→pass | 7,393 | 4,120 | -44% | 1 | 1 | 0% | 1,176 | 1,262 | +7% | 0 | 0 | — |
case-08 | fail→fail | 14,848 | 26,880 | +81% | 1 | 1 | 0% | 3,070 | 5,208 | +70% | 0 | 0 | — |
case-09 | pass→fail | 13,093 | 2,778 | -79% | 1 | 1 | 0% | 1,886 | 944 | -50% | 0 | 0 | — |
case-10 | pass→pass | 12,240 | 7,400 | -40% | 1 | 1 | 0% | 1,998 | 1,729 | -13% | 0 | 0 | — |
case-15 | pass→pass | 12,184 | 8,278 | -32% | 1 | 1 | 0% | 2,144 | 1,896 | -12% | 0 | 0 | — |
case-11 | pass→pass | 10,141 | 3,105 | -69% | 1 | 1 | 0% | 1,501 | 1,049 | -30% | 0 | 0 | — |
case-12 | fail→fail | 14,031 | 5,901 | -58% | 1 | 1 | 0% | 2,435 | 828 | -66% | 0 | 0 | — |
case-13 | fail→pass | 6,209 | 2,325 | -63% | 1 | 1 | 0% | 793 | 884 | +11% | 0 | 0 | — |
case-14 | fail→fail | 14,059 | 7,936 | -44% | 1 | 1 | 0% | 2,285 | 1,893 | -17% | 0 | 0 | — |
case-16 | pass→pass | 6,381 | 3,090 | -52% | 1 | 1 | 0% | 993 | 1,102 | +11% | 0 | 0 | — |
case-17 | pass→pass | 3,680 | 3,002 | -18% | 1 | 1 | 0% | 528 | 993 | +88% | 0 | 0 | — |
case-18 | fail→pass | 14,650 | 6,045 | -59% | 1 | 1 | 0% | 2,198 | 1,495 | -32% | 0 | 0 | — |
case-19 | pass→pass | 10,748 | 2,918 | -73% | 1 | 1 | 0% | 1,677 | 990 | -41% | 0 | 0 | — |
case-20 | fail→fail | 4,662 | 2,251 | -52% | 1 | 1 | 0% | 552 | 846 | +53% | 0 | 0 | — |
case-21 | fail→pass | 3,988 | 7,997 | +101% | 1 | 1 | 0% | 643 | 1,701 | +165% | 0 | 0 | — |
case-22 | pass→pass | 9,601 | 9,274 | -3% | 1 | 1 | 0% | 1,596 | 2,004 | +26% | 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, and 19 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +14 percentage points is the difference between those two pass rates over the 19 comparable cases. 2 cases got worse with the skill loaded, and they are 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.