Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Plan a feature from BRD-lite brief or clear intent into PRD (What), decisions, implementation plan, and task slices.
.claude/skills/hoangnguyen0403-plan-feature/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 156% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 119% | 0% |
> !IMPORTANT] > Plan a feature from BRD-lite brief or clear intent into PRD (What), decisions, implementation plan, and task slices.
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: Produce a PM-owned decision-complete PRD, delivery plan, and IT Department handoff before code starts.
docs/brd/ for the matching [slug]; if multiple candidates exist and intent is unclear, ask the user to choose/input the target slug.common-product-requirements, common-operator-profile, quality-engineering-business-analysis, and matched framework skills.operator_profile, BRD objective IDs, SMART metric, scope fence, assumptions, glossary, risks, and delivery context.docs/prd/prd-[slug].md when file writes are allowed.operator_profile=business, points otherwise.design-solution is required before coding.design-solution; approved build-ready plan with AC IDs and trace -> implementation-readiness.slug, operator_profile, PRD path, REQ-*, AC-*, decisions, RACI, rollout notes, task slices, delivery-window estimate, verification plan, outcome report, next workflow.md# Feature Plan: [Name] ## PRD ## Problem Statement ## Goals And Guardrails ## Personas / JTBD ## Use Cases ## Requirement Trace ## User Stories And ACs ## Decisions | Decision | Choice | Reason | | ---------- | -------- | -------- | | [decision] | [choice] | [reason] | ## RACI / IT Department Handoff ## Analytics / Telemetry ## Risks And Assumptions ## Rollout / Ops ## Implementation Plan ## Task Slices And Delivery Estimate ## Verification Plan ## Outcome Report feature_status: requirements_ready | blocked requirement_trace: BRD-OBJ-* -> REQ-* -> AC-* completed_evidence: []; missing_evidence: []; decision_needed: []; recommended_next_workflow: design-solution | implementation-readiness ## Next Workflow design-solution | implementation-readiness ## Cost Report Call `get_session_cost(workflow="plan-feature")` before final handoff.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→pass | 11,501 | 12,146 | +6% | 1 | 1 | 0% | 1,748 | 2,806 | +61% | 0 | 0 | — |
case-06 | pass→pass | 16,350 | 16,242 | -1% | 1 | 1 | 0% | 2,563 | 3,524 | +37% | 0 | 0 | — |
case-01 | fail→fail | 33,809 | 4,560 | -87% | 1 | 1 | 0% | 6,229 | 1,430 | -77% | 0 | 0 | — |
case-02 | fail→pass | 34,520 | 38,697 | +12% | 1 | 1 | 0% | 6,224 | 7,305 | +17% | 0 | 0 | — |
case-03 | fail→fail | 24,677 | 4,781 | -81% | 1 | 1 | 0% | 4,944 | 1,437 | -71% | 0 | 0 | — |
case-04 | pass→fail | 16,820 | 23,654 | +41% | 1 | 1 | 0% | 2,971 | 5,490 | +85% | 0 | 0 | — |
case-05 | fail→pass | 28,834 | 24,604 | -15% | 1 | 1 | 0% | 5,895 | 5,577 | -5% | 0 | 0 | — |
case-08 | pass→fail | 15,588 | 5,620 | -64% | 1 | 1 | 0% | 2,867 | 1,498 | -48% | 0 | 0 | — |
case-09 | fail→pass | 15,525 | 34,202 | +120% | 1 | 1 | 0% | 2,846 | 7,277 | +156% | 0 | 0 | — |
case-10 | fail→pass | 13,837 | 20,143 | +46% | 1 | 1 | 0% | 2,409 | 5,267 | +119% | 0 | 0 | — |
case-11 | pass→pass | 7,963 | 18,786 | +136% | 1 | 1 | 0% | 1,371 | 4,800 | +250% | 0 | 0 | — |
case-12 | fail→fail | 13,994 | 4,804 | -66% | 1 | 1 | 0% | 2,403 | 1,430 | -40% | 0 | 0 | — |
case-13 | fail→pass | 10,934 | 14,422 | +32% | 1 | 1 | 0% | 1,744 | 3,997 | +129% | 0 | 0 | — |
case-14 | fail→fail | 18,993 | 35,969 | +89% | 1 | 1 | 0% | 3,428 | 7,332 | +114% | 0 | 0 | — |
case-15 | fail→pass | 16,017 | 32,572 | +103% | 1 | 1 | 0% | 3,034 | 6,968 | +130% | 0 | 0 | — |
case-16 | fail→pass | 17,127 | 31,926 | +86% | 1 | 1 | 0% | 3,266 | 7,333 | +125% | 0 | 0 | — |
case-17 | fail→fail | 23,642 | 4,708 | -80% | 1 | 1 | 0% | 4,315 | 1,415 | -67% | 0 | 0 | — |
case-18 | pass→pass | 17,877 | 34,286 | +92% | 1 | 1 | 0% | 3,118 | 7,025 | +125% | 0 | 0 | — |
case-19 | fail→pass | 9,243 | 11,758 | +27% | 1 | 1 | 0% | 1,520 | 2,884 | +90% | 0 | 0 | — |
case-20 | pass→pass | 18,589 | 33,151 | +78% | 1 | 1 | 0% | 3,401 | 7,323 | +115% | 0 | 0 | — |
case-21 | pass→fail | 16,665 | 4,855 | -71% | 1 | 1 | 0% | 3,378 | 1,345 | -60% | 0 | 0 | — |
case-22 | pass→pass | 16,181 | 34,601 | +114% | 1 | 1 | 0% | 2,971 | 7,470 | +151% | 0 | 0 | — |
case-23 | fail→pass | 12,227 | 7,899 | -35% | 1 | 1 | 0% | 2,107 | 2,344 | +11% | 0 | 0 | — |
case-24 | pass→pass | 20,929 | 35,666 | +70% | 1 | 1 | 0% | 3,582 | 7,618 | +113% | 0 | 0 | — |
case-25 | fail→pass | 7,966 | 4,508 | -43% | 1 | 1 | 0% | 1,367 | 1,898 | +39% | 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. 25 cases were attempted, and 19 counted toward the lift figure. The other 6 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 +32 percentage points is the difference between those two pass rates over the 19 comparable cases. 4 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.