Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Route a task to the next synced SDLC workflow based on current artifacts and repo state.
.claude/skills/hoangnguyen0403-sdlc/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 103% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 63% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 111% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 75% | 0% |
> !IMPORTANT] > Route a task to the next synced SDLC workflow based on current artifacts and repo state.
Optional args: slug=<feature>, ticket=<id/url>, mode=interactive|autonomous|channel, channel=<id>, auto_continue=true|false, profile=business|hybrid|technical.
When the user asks to perform this workflow, execute the following steps:
Goal: Select the next native workflow without loading every workflow body, while preserving a traceable BA -> PM -> IT Department handoff for offshore delivery.
operator_profile (business | hybrid | technical) per common-operator-profile and carry it in every Handoff Payload.docs/brd/, docs/prd/, and docs/srs/ for a matching [slug]; if absent, use the newest BRD or git status. Slug = lowercase kebab-case, minted once at brainstorm-feature, reused verbatim downstream; never re-derived.docs/requirements-standards-baseline.mdBRD-OBJ-* -> REQ-* -> AC-* -> SRS-* -> test evidencecommon-task-complexity-routing or the scout SNC: line; label as inference until scouted.recommended_next_workflow, (2) production-incident/urgent-regression signals, (3) earliest missing artifact along the chain below — never skip forward past a gap, (4) cross-cutting audits only on request or as a pre-release gate):brainstorm-featureplan-featuresystem-design-sessiondesign-solutiondesign-solutionimplementation-readinessimplement-featureincident-hotfixdev-fixreview-ticketreview-system-designcode-reviewtest-loop.verify-workverify-bugverify-work PASS on a feature, business acceptance not yet granted -> uat-signoffdeploy-releasepublish-notestraceability-auditsession-reportretro-learncodebase-reviewdesign-solution is required.docs/brd/ and docs/prd/ have no matching [slug] for it yet, the FIRST slice dispatched must be a brainstorm-feature or plan-feature intake slice (owner ba-agent/pm-agent) — never an implementation-readiness/design-solution/implement-feature slice as slice-01. Missing repo roots, OAuth/client IDs, or session-policy decisions are a sign the BRD/PRD step was skipped, not questions to resolve inline in an implementation slice.fast review; medium -> TDD + self-review + deep review; high -> plan-first HARD STOP, independent reviewers, human merge approval.slug, operator_profile, snc_tier, model_tier, recommended workflow, requirement layer, handoff owner, required input, blocking gaps, offshore delivery notes.md# SDLC Route ## Recommended Workflow ## Requirement Layer ## Handoff Owner ## Required Input ## Blocking Gaps ## Offshore Delivery Notes ## Outcome Report feature_status: not_started | requirements_ready | design_ready | partially_implemented | implemented | blocked requirement_trace: BRD-OBJ-* -> REQ-* -> AC-* -> SRS-* -> evidence completed_evidence: []; missing_evidence: []; decision_needed: []; recommended_next_workflow: ## Next Workflow ## Verification Command ## Cost Report Call `get_session_cost(workflow="sdlc")` before final handoff.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 14,812 | 15,246 | +3% | 1 | 1 | 0% | 376 | 1,999 | +432% | 0 | 0 | — |
case-02 | pass→pass | 24,426 | 24,757 | +1% | 1 | 1 | 0% | 3,254 | 5,333 | +64% | 0 | 0 | — |
case-03 | fail→fail | 19,527 | 15,513 | -21% | 1 | 1 | 0% | 2,986 | 2,009 | -33% | 0 | 0 | — |
case-04 | fail→pass | 17,954 | 19,112 | +6% | 1 | 1 | 0% | 2,151 | 4,368 | +103% | 0 | 0 | — |
case-05 | fail→pass | 19,428 | 16,104 | -17% | 1 | 1 | 0% | 2,415 | 3,941 | +63% | 0 | 0 | — |
case-06 | pass→pass | 20,548 | 19,589 | -5% | 1 | 1 | 0% | 2,784 | 4,613 | +66% | 0 | 0 | — |
case-07 | fail→pass | 19,069 | 13,196 | -31% | 1 | 1 | 0% | 2,565 | 3,237 | +26% | 0 | 0 | — |
case-08 | fail→fail | 16,519 | 17,843 | +8% | 1 | 1 | 0% | 1,958 | 1,959 | +0% | 0 | 0 | — |
case-09 | fail→pass | 14,334 | 13,382 | -7% | 1 | 1 | 0% | 1,554 | 3,277 | +111% | 0 | 0 | — |
case-10 | pass→pass | 12,913 | 8,919 | -31% | 1 | 1 | 0% | 1,436 | 3,250 | +126% | 0 | 0 | — |
case-11 | fail→fail | 10,769 | 12,146 | +13% | 1 | 1 | 0% | 1,013 | 3,093 | +205% | 0 | 0 | — |
case-12 | fail→pass | 16,507 | 14,255 | -14% | 1 | 1 | 0% | 1,966 | 3,439 | +75% | 0 | 0 | — |
case-13 | pass→pass | 20,719 | 14,256 | -31% | 1 | 1 | 0% | 2,874 | 3,580 | +25% | 0 | 0 | — |
case-14 | fail→pass | 28,792 | 26,910 | -7% | 1 | 1 | 0% | 3,797 | 3,606 | -5% | 0 | 0 | — |
case-15 | fail→pass | 28,986 | 16,198 | -44% | 1 | 1 | 0% | 4,521 | 3,557 | -21% | 0 | 0 | — |
case-16 | pass→pass | 31,374 | 7,936 | -75% | 1 | 1 | 0% | 4,141 | 3,019 | -27% | 0 | 0 | — |
case-17 | fail→pass | 19,586 | 28,335 | +45% | 1 | 1 | 0% | 2,659 | 4,680 | +76% | 0 | 0 | — |
case-18 | pass→pass | 16,807 | 13,123 | -22% | 1 | 1 | 0% | 1,948 | 3,244 | +67% | 0 | 0 | — |
case-19 | pass→pass | 18,392 | 16,386 | -11% | 1 | 1 | 0% | 2,284 | 3,744 | +64% | 0 | 0 | — |
case-20 | fail→pass | 12,319 | 18,882 | +53% | 1 | 1 | 0% | 1,255 | 4,313 | +244% | 0 | 0 | — |
case-21 | fail→pass | 17,674 | 17,586 | -0% | 1 | 1 | 0% | 2,125 | 4,215 | +98% | 0 | 0 | — |
case-22 | pass→fail | 17,486 | 25,874 | +48% | 1 | 1 | 0% | 2,566 | 4,375 | +70% | 0 | 0 | — |
case-23 | pass→fail | 16,730 | 21,888 | +31% | 1 | 1 | 0% | 2,123 | 4,987 | +135% | 0 | 0 | — |
case-24 | pass→fail | 18,140 | 15,746 | -13% | 1 | 1 | 0% | 2,309 | 1,956 | -15% | 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 20 counted toward the lift figure. The other 4 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 +29 percentage points is the difference between those two pass rates over the 20 comparable cases. 4 cases got worse with the skill loaded, and they are included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v2, not the version currently listed.
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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/2/2026 | +44% |
Other measured skills in the registry, with their headline benchmark lift.