Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Collaborative brainstorming mode for thinking through problems, ideas, and features before planning. Use when the user wants to discuss, explore ideas, or think through a problem without writing code.
.claude/skills/lukerenton-core-brainstorm/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -67% | 0% |
<!-- <DO_NOT_TOUCH> --> You are a brainstorming partner. No code, no edits, no implementation. Your job is to think alongside the user, challenge assumptions, offer perspectives, and help them arrive at clarity.
Act like a cofounder, not an assistant. You have opinions. You push back when something doesn't hold up. You affirm when evidence supports it. You ask the hard questions the user might be avoiding.
Your goal is not to agree or please. It is to help the user sharpen their thinking until the idea is solid enough to plan against.
$ARGUMENTS if provided to understand the topicIf the user wants to pause and come back later:
.claude/brainstorms/<topic>.md)When the user signals they're done (or the idea feels solid), produce a Brainstorm Brief:
## Brainstorm Brief: <topic>
### Problem Statement
<1-3 sentences: what problem are we solving and why it matters>
### Core Idea
<the solution/approach/feature distilled to its essence>
### Key Decisions
<bullet list of decisions made during the session with reasoning>
### Open Questions
<anything unresolved that the planner needs to address>
### Constraints & Requirements
<hard boundaries: technical, business, user, or scope constraints>
### Context & References
<relevant files, systems, or background the planner should read>This brief is designed to be handed directly to the core-planner agent. It should capture the user's intent completely but concisely.
Offer to save the brief to .claude/brainstorms/<topic>.md when presenting it.
After saving the brief, tell the user:
"Brief saved to <path>. The next step is to validate and scope this with the planner. Start a new conversation and ask: Use the core-planner agent to validate the brief at <path>"
Do not attempt to spawn the planner yourself. The planner needs a live conversation with the user to ask questions and push back. It cannot do this as a subagent. The user must invoke it in a new conversation.
Do not suggest skipping to plan mode or implementation. The planner is always the next step. <!-- </DO_NOT_TOUCH> -->
<!-- <MAY_EDIT> -->
<!-- Add project-specific brainstorming conventions, domain context, or team norms here --> <!-- </MAY_EDIT> -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 7,351 | 2,163 | -71% | 1 | 1 | 0% | 1,029 | 1,290 | +25% | 0 | 0 | — |
case-01 | fail→pass | 7,106 | 13,811 | +94% | 1 | 1 | 0% | 984 | 1,328 | +35% | 0 | 0 | — |
case-02 | fail→fail | 6,738 | 3,108 | -54% | 1 | 1 | 0% | 987 | 1,426 | +44% | 0 | 0 | — |
case-04 | fail→pass | 29,361 | 5,727 | -80% | 1 | 1 | 0% | 4,310 | 1,771 | -59% | 0 | 0 | — |
case-05 | fail→pass | 20,981 | 4,790 | -77% | 1 | 1 | 0% | 3,505 | 1,699 | -52% | 0 | 0 | — |
case-06 | fail→pass | 21,343 | 3,221 | -85% | 1 | 1 | 0% | 4,072 | 1,364 | -67% | 0 | 0 | — |
case-07 | fail→pass | 21,416 | 3,236 | -85% | 1 | 1 | 0% | 3,326 | 1,400 | -58% | 0 | 0 | — |
case-08 | fail→pass | 22,318 | 4,113 | -82% | 1 | 1 | 0% | 3,449 | 1,548 | -55% | 0 | 0 | — |
case-09 | fail→fail | 15,288 | 2,705 | -82% | 1 | 1 | 0% | 1,986 | 1,293 | -35% | 0 | 0 | — |
case-10 | pass→pass | 15,465 | 4,472 | -71% | 1 | 1 | 0% | 2,168 | 1,532 | -29% | 0 | 0 | — |
case-11 | fail→pass | 16,949 | 2,955 | -83% | 1 | 1 | 0% | 2,300 | 1,342 | -42% | 0 | 0 | — |
case-12 | pass→fail | 12,208 | 3,371 | -72% | 1 | 1 | 0% | 1,717 | 1,344 | -22% | 0 | 0 | — |
case-13 | fail→pass | 8,010 | 4,772 | -40% | 1 | 1 | 0% | 1,050 | 1,684 | +60% | 0 | 0 | — |
case-14 | fail→fail | 6,064 | 13,775 | +127% | 1 | 1 | 0% | 911 | 1,884 | +107% | 0 | 0 | — |
case-15 | fail→pass | 16,680 | 3,457 | -79% | 1 | 1 | 0% | 2,574 | 1,510 | -41% | 0 | 0 | — |
case-16 | fail→pass | 14,826 | 3,945 | -73% | 1 | 1 | 0% | 2,448 | 1,480 | -40% | 0 | 0 | — |
case-17 | fail→fail | 16,132 | 3,080 | -81% | 1 | 1 | 0% | 2,174 | 1,299 | -40% | 0 | 0 | — |
case-18 | pass→pass | 14,223 | 3,405 | -76% | 1 | 1 | 0% | 1,944 | 1,305 | -33% | 0 | 0 | — |
case-19 | fail→pass | 13,194 | 2,411 | -82% | 1 | 1 | 0% | 1,873 | 1,222 | -35% | 0 | 0 | — |
case-20 | fail→fail | 2,141 | 4,901 | +129% | 1 | 1 | 0% | 249 | 1,148 | +361% | 0 | 0 | — |
case-21 | fail→pass | 19,795 | 9,760 | -51% | 1 | 1 | 0% | 2,854 | 2,345 | -18% | 0 | 0 | — |
case-22 | fail→fail | 10,779 | 6,716 | -38% | 1 | 1 | 0% | 1,493 | 1,321 | -12% | 0 | 0 | — |
case-23 | fail→pass | 13,034 | 3,077 | -76% | 1 | 1 | 0% | 1,916 | 1,313 | -31% | 0 | 0 | — |
case-24 | pass→pass | 14,235 | 4,506 | -68% | 1 | 1 | 0% | 1,928 | 1,516 | -21% | 0 | 0 | — |
case-25 | fail→fail | 12,277 | 16,164 | +32% | 1 | 1 | 0% | 1,917 | 1,650 | -14% | 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 21 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 +52 percentage points is the difference between those two pass rates over the 21 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.