Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Clarify a rough product or engineering idea into a BRD-lite brief (Why) with measurable business value.
.claude/skills/hoangnguyen0403-brainstorm-feature/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 83% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 33% | 0% |
> !IMPORTANT] > Clarify a rough product or engineering idea into a BRD-lite brief (Why) with measurable business value.
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: Convert vague intent into a compact BA-owned BRD-lite brief before PM PRD planning or technical design.
common-business-requirements, and common-operator-profile.operator_profile (business | hybrid | technical) from request phrasing; never ask the operator to self-rate. Carry it in the Handoff Payload.REQ-* placeholders and flag platform, market, permission, and edge-case gaps for PM.assumed); scope fence drafted from the request with explicit non-goals.operator_profile=business, present all three drafted defaults as one confirm-with-default question round (fits the max-3 rule) instead of blocking outright.docs/brd/brd-[slug].md when writes are allowed and route to plan-feature.slug, operator_profile, executive summary, business objective, SMART metric, recommended approach, alternatives, constraints, non-goals, open questions, assumptions (flagged assumed), PM handoff checklist.feature_status=requirements_ready | blocked, requirement trace seed, completed/missing evidence, decision needed, and recommended next workflow.md# BRD-lite Brief: [Name] ## Executive Summary ## Business Objective ## SMART Success Metric ## Target Users ## Problem ## AS-IS To TO-BE ## Stakeholders And Validation Owner ## Success Metrics ## Cost-Benefit / Value Hypothesis ## Offshore Delivery Context ## Recommended Approach ## Alternatives Considered ## Stakeholders ## Constraints ## Non-Goals ## Glossary ## PM Handoff Checklist ## Outcome Report feature_status: requirements_ready | blocked requirement_trace: BRD-OBJ-* -> candidate REQ-* completed_evidence: []; missing_evidence: []; decision_needed: []; assumptions: []; recommended_next_workflow: plan-feature ## Open Questions ## Next Workflow plan-feature ## Cost Report Call `get_session_cost(workflow="brainstorm-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-10 | fail→pass | 14,744 | 16,867 | +14% | 1 | 1 | 0% | 2,452 | 3,823 | +56% | 0 | 0 | — |
case-11 | fail→pass | 15,850 | 21,042 | +33% | 1 | 1 | 0% | 2,787 | 4,359 | +56% | 0 | 0 | — |
case-01 | fail→pass | 18,273 | 19,029 | +4% | 1 | 1 | 0% | 3,166 | 4,316 | +36% | 0 | 0 | — |
case-02 | fail→pass | 16,085 | 22,760 | +41% | 1 | 1 | 0% | 2,786 | 5,089 | +83% | 0 | 0 | — |
case-03 | fail→pass | 19,171 | 18,726 | -2% | 1 | 1 | 0% | 3,294 | 4,372 | +33% | 0 | 0 | — |
case-04 | pass→fail | 29,086 | 28,127 | -3% | 1 | 1 | 0% | 6,193 | 6,275 | +1% | 0 | 0 | — |
case-05 | pass→fail | 30,093 | 24,860 | -17% | 1 | 1 | 0% | 6,180 | 5,450 | -12% | 0 | 0 | — |
case-06 | pass→pass | 18,925 | 26,497 | +40% | 1 | 1 | 0% | 3,720 | 5,951 | +60% | 0 | 0 | — |
case-07 | fail→pass | 13,113 | 21,771 | +66% | 1 | 1 | 0% | 2,359 | 5,212 | +121% | 0 | 0 | — |
case-08 | pass→pass | 18,475 | 17,315 | -6% | 1 | 1 | 0% | 3,045 | 3,918 | +29% | 0 | 0 | — |
case-09 | fail→pass | 7,472 | 11,292 | +51% | 1 | 1 | 0% | 1,251 | 2,957 | +136% | 0 | 0 | — |
case-12 | fail→pass | 16,006 | 20,681 | +29% | 1 | 1 | 0% | 2,695 | 4,529 | +68% | 0 | 0 | — |
case-13 | pass→pass | 9,774 | 6,587 | -33% | 1 | 1 | 0% | 1,655 | 2,087 | +26% | 0 | 0 | — |
case-14 | pass→pass | 13,484 | 18,975 | +41% | 1 | 1 | 0% | 2,363 | 4,383 | +85% | 0 | 0 | — |
case-15 | fail→pass | 10,431 | 10,504 | +1% | 1 | 1 | 0% | 1,709 | 2,743 | +61% | 0 | 0 | — |
case-16 | fail→pass | 7,525 | 5,429 | -28% | 1 | 1 | 0% | 1,152 | 1,850 | +61% | 0 | 0 | — |
case-17 | fail→pass | 13,395 | 19,640 | +47% | 1 | 1 | 0% | 2,308 | 4,626 | +100% | 0 | 0 | — |
case-18 | fail→pass | 10,895 | 16,118 | +48% | 1 | 1 | 0% | 1,819 | 3,364 | +85% | 0 | 0 | — |
case-19 | fail→fail | 13,835 | 23,523 | +70% | 1 | 1 | 0% | 2,448 | 5,185 | +112% | 0 | 0 | — |
case-20 | pass→pass | 10,472 | 6,094 | -42% | 1 | 1 | 0% | 1,675 | 1,962 | +17% | 0 | 0 | — |
case-21 | fail→pass | 17,226 | 9,703 | -44% | 1 | 1 | 0% | 2,651 | 2,569 | -3% | 0 | 0 | — |
case-22 | fail→fail | 12,168 | 15,435 | +27% | 1 | 1 | 0% | 2,452 | 3,733 | +52% | 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 +50 percentage points is the difference between those two pass rates over the 22 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.