Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Fullstack engineering review — walks the 7 Matt Pocock forcing questions, picks the profile, forks into POWERFUL specialists (api-design-reviewer, database-designer, slo-architect). Invokes the cs-fullstack-engineer agent with context fork.
.claude/skills/alirezarezvani-csfullstack-review-fullstack-engineering-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -75% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -50% | 0% |
Use the cs-fullstack-engineer agent (which uses context: fork to keep the parent thread clean) to handle this inquiry:
$ARGUMENTS
Canonical source: engineering-team/skills/senior-fullstack/references/forcing_questions.md (7 questions, one-per-turn, recommendation + canon citation per question).
engineering-team/skills/senior-fullstack/references/forcing_questions.md. One per turn. Recommend the answer with cited canon. Track in /tmp/fullstack-grill-<date>.md.bash python engineering-team/skills/senior-fullstack/scripts/fullstack_decision_engine.py \ --team-size <N> --team-size-12mo <N12> --cadence <c> \ --user-facing <true|false> --budget <USD/mo> \ --traffic-p99-rps <N> --data-sensitivity <tier>
api-design-reviewer for API contractdatabase-designer for schemaslo-architect for reliability targetci-cd-pipeline-builder for the pipelineperformance-profiler for perf baselinecs-karpathy-reviewer before any commitProfiles live at engineering-team/skills/senior-fullstack/profiles/. To customize for your org:
saas-startup.json (or whichever best fits) to <your-org>.json.constraints, stack_recommendations, success_thresholds, named_approver_chain./cs:frontend-review — frontend-only deep dive/cs:backend-review — backend-only deep dive/cs:engineer-grill — cross-role 21-question forcing-question runner/karpathy-check — Karpathy 4-principle review before commit| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 8,058 | 4,996 | -38% | 1 | 1 | 0% | 1,193 | 1,585 | +33% | 0 | 0 | — |
case-02 | fail→pass | 40,271 | 8,524 | -79% | 1 | 1 | 0% | 6,211 | 1,527 | -75% | 0 | 0 | — |
case-03 | fail→pass | 5,865 | 4,979 | -15% | 1 | 1 | 0% | 1,014 | 1,568 | +55% | 0 | 0 | — |
case-04 | fail→pass | 5,742 | 3,591 | -37% | 1 | 1 | 0% | 897 | 1,306 | +46% | 0 | 0 | — |
case-05 | fail→pass | 15,539 | 3,476 | -78% | 1 | 1 | 0% | 2,772 | 1,392 | -50% | 0 | 0 | — |
case-06 | fail→pass | 13,404 | 5,360 | -60% | 1 | 1 | 0% | 2,417 | 1,557 | -36% | 0 | 0 | — |
case-07 | fail→pass | 11,492 | 9,944 | -13% | 1 | 1 | 0% | 1,638 | 1,748 | +7% | 0 | 0 | — |
case-08 | fail→pass | 11,857 | 4,038 | -66% | 1 | 1 | 0% | 1,795 | 1,622 | -10% | 0 | 0 | — |
case-09 | fail→pass | 10,462 | 7,004 | -33% | 1 | 1 | 0% | 1,672 | 2,052 | +23% | 0 | 0 | — |
case-10 | fail→pass | 29,230 | 3,442 | -88% | 1 | 1 | 0% | 2,972 | 1,433 | -52% | 0 | 0 | — |
case-11 | fail→pass | 14,685 | 6,606 | -55% | 1 | 1 | 0% | 2,421 | 1,859 | -23% | 0 | 0 | — |
case-12 | fail→pass | 6,127 | 6,470 | +6% | 1 | 1 | 0% | 1,106 | 1,805 | +63% | 0 | 0 | — |
case-13 | fail→pass | 11,282 | 2,863 | -75% | 1 | 1 | 0% | 1,776 | 1,168 | -34% | 0 | 0 | — |
case-14 | fail→pass | 6,350 | 2,795 | -56% | 1 | 1 | 0% | 1,037 | 1,274 | +23% | 0 | 0 | — |
case-15 | fail→pass | 8,544 | 4,456 | -48% | 1 | 1 | 0% | 1,761 | 1,460 | -17% | 0 | 0 | — |
case-16 | fail→pass | 8,123 | 1,941 | -76% | 1 | 1 | 0% | 1,446 | 1,123 | -22% | 0 | 0 | — |
case-17 | fail→pass | 5,767 | 4,479 | -22% | 1 | 1 | 0% | 768 | 1,452 | +89% | 0 | 0 | — |
case-18 | fail→pass | 9,950 | 2,372 | -76% | 1 | 1 | 0% | 1,810 | 1,192 | -34% | 0 | 0 | — |
case-19 | fail→pass | 12,966 | 3,400 | -74% | 1 | 1 | 0% | 2,455 | 1,458 | -41% | 0 | 0 | — |
case-20 | pass→pass | 17,093 | 10,796 | -37% | 1 | 1 | 0% | 2,885 | 2,969 | +3% | 0 | 0 | — |
case-21 | pass→pass | 11,277 | 6,652 | -41% | 1 | 1 | 0% | 2,414 | 2,124 | -12% | 0 | 0 | — |
case-22 | pass→pass | 6,224 | 5,844 | -6% | 1 | 1 | 0% | 1,323 | 1,677 | +27% | 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 +86 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.