Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when completing tasks, implementing major features, or before merging to verify work meets requirements
.claude/skills/ibrahim-3d-requesting-code-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-14 | ✓→✗ | ▼ Worse | 58% | 0% |
| case-13 | ✓→✓ | = Same ✓ | 0% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 50% | 0% |
Dispatch orchestrator-supaconductor:code-reviewer subagent to catch issues before they cascade.
Core principle: Review early, review often.
Mandatory:
Optional but valuable:
1. Get git SHAs:
bashBASE_SHA=$(git rev-parse HEAD~1) # or origin/main HEAD_SHA=$(git rev-parse HEAD)
2. Dispatch code-reviewer subagent:
Use Task tool with orchestrator-supaconductor:code-reviewer type, fill template at code-reviewer.md
Placeholders:
{WHAT_WAS_IMPLEMENTED} - What you just built{PLAN_OR_REQUIREMENTS} - What it should do{BASE_SHA} - Starting commit{HEAD_SHA} - Ending commit{DESCRIPTION} - Brief summary3. Act on feedback:
[Just completed Task 2: Add verification function]
You: Let me request code review before proceeding.
BASE_SHA=$(git log --oneline | grep_search "Task 1" | head -1 | awk '{print $1}')
HEAD_SHA=$(git rev-parse HEAD)
[Dispatch orchestrator-supaconductor:code-reviewer subagent]
WHAT_WAS_IMPLEMENTED: Verification and repair functions for conversation index
PLAN_OR_REQUIREMENTS: Task 2 from conductor/tracks/deployment_20260327/plan.md
BASE_SHA: a7981ec
HEAD_SHA: 3df7661
DESCRIPTION: Added verifyIndex() and repairIndex() with 4 issue types
[Subagent returns]:
Strengths: Clean architecture, real tests
Issues:
Important: Missing progress indicators
Minor: Magic number (100) for reporting interval
Assessment: Ready to proceed
You: [Fix progress indicators]
[Continue to Task 3]Subagent-Driven Development:
Executing Plans:
Ad-Hoc Development:
Never:
If reviewer wrong:
See template at: requesting-code-review/code-reviewer.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | pass→pass | 6,374 | 2,902 | -54% | 1 | 1 | 0% | 1,158 | 1,158 | 0% | 0 | 0 | — |
case-01 | fail→fail | 3,524 | 12,146 | +245% | 1 | 1 | 0% | 571 | 1,269 | +122% | 0 | 0 | — |
case-02 | fail→fail | 5,403 | 5,752 | +6% | 1 | 1 | 0% | 781 | 1,008 | +29% | 0 | 0 | — |
case-03 | fail→fail | 6,001 | 4,624 | -23% | 1 | 1 | 0% | 922 | 892 | -3% | 0 | 0 | — |
case-04 | pass→pass | 12,713 | 14,265 | +12% | 1 | 1 | 0% | 2,132 | 3,194 | +50% | 0 | 0 | — |
case-05 | pass→pass | 9,567 | 7,257 | -24% | 1 | 1 | 0% | 1,648 | 2,007 | +22% | 0 | 0 | — |
case-06 | pass→pass | 4,797 | 3,416 | -29% | 1 | 1 | 0% | 835 | 1,231 | +47% | 0 | 0 | — |
case-07 | pass→pass | 9,268 | 5,031 | -46% | 1 | 1 | 0% | 1,396 | 1,498 | +7% | 0 | 0 | — |
case-08 | pass→pass | 23,585 | 3,311 | -86% | 1 | 1 | 0% | 1,367 | 1,291 | -6% | 0 | 0 | — |
case-09 | pass→pass | 5,304 | 3,895 | -27% | 1 | 1 | 0% | 808 | 1,285 | +59% | 0 | 0 | — |
case-10 | pass→pass | 9,870 | 8,037 | -19% | 1 | 1 | 0% | 1,520 | 1,893 | +25% | 0 | 0 | — |
case-11 | pass→pass | 10,348 | 3,537 | -66% | 1 | 1 | 0% | 1,606 | 1,254 | -22% | 0 | 0 | — |
case-12 | fail→pass | 12,840 | 2,710 | -79% | 1 | 1 | 0% | 2,090 | 1,112 | -47% | 0 | 0 | — |
case-14 | pass→fail | 5,909 | 4,797 | -19% | 1 | 1 | 0% | 952 | 1,504 | +58% | 0 | 0 | — |
case-15 | fail→fail | 25,646 | 4,303 | -83% | 1 | 1 | 0% | 914 | 1,497 | +64% | 0 | 0 | — |
case-16 | pass→pass | 4,389 | 3,211 | -27% | 1 | 1 | 0% | 656 | 1,181 | +80% | 0 | 0 | — |
case-17 | pass→pass | 10,738 | 4,075 | -62% | 1 | 1 | 0% | 1,588 | 1,343 | -15% | 0 | 0 | — |
case-18 | pass→pass | 12,172 | 6,994 | -43% | 1 | 1 | 0% | 1,847 | 1,893 | +2% | 0 | 0 | — |
case-19 | pass→pass | 4,758 | 4,079 | -14% | 1 | 1 | 0% | 752 | 1,262 | +68% | 0 | 0 | — |
case-20 | fail→pass | 15,161 | 2,014 | -87% | 1 | 1 | 0% | 2,600 | 993 | -62% | 0 | 0 | — |
case-21 | pass→pass | 10,001 | 3,082 | -69% | 1 | 1 | 0% | 1,575 | 1,177 | -25% | 0 | 0 | — |
case-22 | pass→pass | 10,571 | 5,558 | -47% | 1 | 1 | 0% | 1,660 | 1,605 | -3% | 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 +5 percentage points is the difference between those two pass rates over the 19 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.