Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Executes automated project-wide refactoring with a focus on cross-file optimization. This skill should be used when the user asks for project-wide refactoring, cross-file simplification, consistency standardization across the codebase, duplication reduction, or invokes "/refactor-project".
.claude/skills/fradser-refactor-project/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -41% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -47% | 0% |
Execute automated project-wide refactoring using refactor:code-simplifier agent with cross-file optimization focus.
Goal: Ensure project-wide execution is explicit and reproducible.
Actions:
git rev-parse --is-inside-work-tree; if false, inform user that project-wide mode requires a git workspacegit rev-parse --short HEAD and include it in final summary for rollback contextGoal: Discover all code files and display scope summary.
Actions:
node_modules/, .git/, dist/, build/, vendor/, .venv/See references/scope-analysis.md for exclusion patterns and edge cases.
Goal: Execute refactor:code-simplifier agent with project-wide scope and cross-file focus.
Actions:
refactor:code-simplifier agent with all discovered code filesrefactor:best-practices skill and applies language-specific patternsSee references/agent-configuration.md for detailed Task parameters.
Goal: Report comprehensive summary of project-wide changes.
Actions:
git restore --worktree --staged .)See references/output-requirements.md for detailed summary format.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 13,038 | 4,316 | -67% | 1 | 1 | 0% | 2,574 | 1,207 | -53% | 0 | 0 | — |
case-01 | fail→fail | 7,084 | 6,871 | -3% | 1 | 1 | 0% | 1,089 | 1,012 | -7% | 0 | 0 | — |
case-02 | fail→fail | 11,725 | 6,334 | -46% | 1 | 1 | 0% | 1,922 | 968 | -50% | 0 | 0 | — |
case-03 | fail→fail | 5,454 | 7,190 | +32% | 1 | 1 | 0% | 248 | 1,022 | +312% | 0 | 0 | — |
case-05 | pass→fail | 12,478 | 4,500 | -64% | 1 | 1 | 0% | 2,238 | 1,196 | -47% | 0 | 0 | — |
case-06 | fail→pass | 12,180 | 3,982 | -67% | 1 | 1 | 0% | 1,768 | 1,136 | -36% | 0 | 0 | — |
case-07 | pass→pass | 12,253 | 2,154 | -82% | 1 | 1 | 0% | 2,031 | 850 | -58% | 0 | 0 | — |
case-08 | fail→pass | 13,122 | 7,397 | -44% | 1 | 1 | 0% | 2,181 | 1,885 | -14% | 0 | 0 | — |
case-09 | fail→fail | 10,510 | 2,441 | -77% | 1 | 1 | 0% | 1,616 | 970 | -40% | 0 | 0 | — |
case-10 | fail→pass | 9,841 | 4,264 | -57% | 1 | 1 | 0% | 1,644 | 966 | -41% | 0 | 0 | — |
case-11 | pass→pass | 14,371 | 5,237 | -64% | 1 | 1 | 0% | 2,108 | 1,332 | -37% | 0 | 0 | — |
case-12 | fail→pass | 17,223 | 10,042 | -42% | 1 | 1 | 0% | 2,687 | 1,793 | -33% | 0 | 0 | — |
case-13 | fail→pass | 12,197 | 2,625 | -78% | 1 | 1 | 0% | 1,752 | 924 | -47% | 0 | 0 | — |
case-14 | pass→pass | 14,073 | 2,700 | -81% | 1 | 1 | 0% | 2,331 | 949 | -59% | 0 | 0 | — |
case-15 | pass→pass | 11,504 | 4,056 | -65% | 1 | 1 | 0% | 1,820 | 1,093 | -40% | 0 | 0 | — |
case-16 | pass→pass | 15,203 | 7,354 | -52% | 1 | 1 | 0% | 2,264 | 1,681 | -26% | 0 | 0 | — |
case-17 | fail→fail | 15,475 | 4,992 | -68% | 1 | 1 | 0% | 2,395 | 1,455 | -39% | 0 | 0 | — |
case-18 | pass→pass | 9,205 | 2,869 | -69% | 1 | 1 | 0% | 1,465 | 935 | -36% | 0 | 0 | — |
case-19 | fail→pass | 15,076 | 5,566 | -63% | 1 | 1 | 0% | 2,348 | 1,563 | -33% | 0 | 0 | — |
case-20 | fail→pass | 6,439 | 2,975 | -54% | 1 | 1 | 0% | 1,162 | 937 | -19% | 0 | 0 | — |
case-21 | pass→pass | 13,349 | 7,530 | -44% | 1 | 1 | 0% | 1,921 | 1,769 | -8% | 0 | 0 | — |
case-22 | pass→pass | 7,871 | 6,263 | -20% | 1 | 1 | 0% | 1,192 | 1,428 | +20% | 0 | 0 | — |
case-23 | pass→fail | 10,611 | 30,023 | +183% | 1 | 1 | 0% | 2,168 | 6,107 | +182% | 0 | 0 | — |
case-24 | fail→fail | 9,061 | 10,271 | +13% | 1 | 1 | 0% | 1,043 | 1,491 | +43% | 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. 24 cases were attempted, and 21 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 +21 percentage points is the difference between those two pass rates over the 21 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.