Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write or update the project's durable design brief at .ui-craft/brief.md. Invoke when the user asks for brief on their UI, or mentions 'brief' alongside design / UI / frontend work.
.claude/skills/educlopez-brief/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-14 | ✓→✗ | ▼ Worse | -31% | 0% |
| case-19 | ✓→✗ | ▼ Worse | -12% | 0% |
| case-20 | ✓→✗ | ▼ Worse | -50% | 0% |
| case-21 | ✓→✗ | ▼ Worse | -45% | 0% |
<!-- HARNESS MIRROR — do not edit here. Canonical source: skills/ or commands/. After editing source, copy into cli/assets/<harness>/ and repo-root harness mirrors. -->
Context: this sub-skill is one lens of the broader ui-craft skill. If the ui-craft skill is also installed, read its SKILL.md first for Discovery + Anti-Slop + Craft Test, then apply the specific lens below.
Load references/brief.md for the brief format before proceeding.
Check whether .ui-craft/brief.md exists. Use Read or ls on .ui-craft/.
If it exists: load its contents. If $ARGUMENTS names a specific update ("update principles", "shift target user", "add out of scope"), focus the session there and skip unchanged sections. Otherwise summarize what's in the brief and ask: "What do you want to change?"
If it doesn't exist: proceed to Step 2. Also check the repo root for a DESIGN.md or design-tokens.json (an ecosystem convention some teams already maintain) — if present, read it and pre-fill the brief's answers from it instead of re-asking; note the source. The brief complements an existing design contract, never contradicts it.
Walk the user through the five required sections in a single pass. Ask ONE compact question per section. Do not open five separate prompts unless the user asks for depth.
Compact prompt template:
> "To write your design brief I need five things — answer as tersely as you like, I'll fill in structure: > 1. Product purpose — what does it do, in one sentence? > 2. Primary user — who, by role and context? > 3. Principles — what does this product believe? Give me three to five stances it takes. (If you don't have these yet, say so — I'll run the principles workshop.) > 4. Success metric — what does 'the user succeeded on this surface' look like in observable behavior? > 5. Out of scope — three to five things this surface deliberately does NOT do."
Principles workshop case: if the user says "I don't have principles yet" or gives vague answers ("make it good", "keep it simple"), do NOT fabricate principles from vague input. Load references/principles-catalog.md first. The catalog has 42 worked example principles across 8 product categories — use them as conversation seeds, not as templates to adopt literally. Show 2-3 from the closest category to the user's product, then ask which resonate or which they'd flip. Then run the principles workshop from references/brief.md as a focused sub-flow: ask for three past design decisions that were debated, then derive candidate principles from the patterns. Refuse to accept platitudes — push back and prompt for substance.
Parsing long input: if the user provides a product description or PRD, parse it into the five sections rather than re-asking for information already given. State what you extracted and ask for confirmation or corrections.
Thin answers: if an answer is too vague to constrain a design decision, ask one targeted follow-up. Maximum two follow-ups per section before flagging it as incomplete and moving on.
Always show the proposed brief in full before writing the file. Ask: "Does this look right, or anything to adjust?"
Do not write until confirmed.
Create .ui-craft/ if it doesn't exist. Write the confirmed content to .ui-craft/brief.md.
The file must contain all five sections with the exact headings from references/brief.md. No omissions.
Tell the user the file is at .ui-craft/brief.md. Suggest committing it:
> "Commit .ui-craft/brief.md to the repo — it's documentation, not config. The agent reads it every session."
Do not auto-commit. Per project rules, commits require explicit user instruction.
[TO DEFINE — YYYY-MM-DD] so the gap is visible.Next step: /tokens — the brief decides the intent, the tokens encode it (rung 2).
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 5,124 | 8,771 | +71% | 1 | 1 | 0% | 259 | 1,230 | +375% | 0 | 0 | — |
case-02 | fail→fail | 9,805 | 3,525 | -64% | 1 | 1 | 0% | 1,479 | 1,457 | -1% | 0 | 0 | — |
case-03 | fail→fail | 2,087 | 4,888 | +134% | 1 | 1 | 0% | 309 | 1,262 | +308% | 0 | 0 | — |
case-04 | fail→fail | 6,134 | 9,611 | +57% | 1 | 1 | 0% | 374 | 1,764 | +372% | 0 | 0 | — |
case-05 | fail→fail | 13,320 | 5,473 | -59% | 1 | 1 | 0% | 2,053 | 1,361 | -34% | 0 | 0 | — |
case-06 | fail→fail | 9,764 | 7,897 | -19% | 1 | 1 | 0% | 1,495 | 1,576 | +5% | 0 | 0 | — |
case-07 | fail→fail | 12,325 | 6,965 | -43% | 1 | 1 | 0% | 2,001 | 1,573 | -21% | 0 | 0 | — |
case-08 | pass→pass | 8,835 | 17,370 | +97% | 1 | 1 | 0% | 1,424 | 2,984 | +110% | 0 | 0 | — |
case-09 | fail→fail | 13,904 | 3,748 | -73% | 1 | 1 | 0% | 2,005 | 1,653 | -18% | 0 | 0 | — |
case-10 | fail→fail | 17,652 | 22,618 | +28% | 1 | 1 | 0% | 2,450 | 4,634 | +89% | 0 | 0 | — |
case-11 | fail→fail | 12,452 | 6,334 | -49% | 1 | 1 | 0% | 2,142 | 1,379 | -36% | 0 | 0 | — |
case-12 | fail→fail | 6,245 | 51,586 | +726% | 1 | 1 | 0% | 520 | 1,710 | +229% | 0 | 0 | — |
case-13 | fail→pass | 6,547 | 5,221 | -20% | 1 | 1 | 0% | 1,057 | 1,865 | +76% | 0 | 0 | — |
case-14 | pass→fail | 10,561 | 4,084 | -61% | 1 | 1 | 0% | 1,776 | 1,222 | -31% | 0 | 0 | — |
case-15 | fail→fail | 7,159 | 6,902 | -4% | 1 | 1 | 0% | 1,095 | 1,623 | +48% | 0 | 0 | — |
case-16 | fail→fail | 19,331 | 6,814 | -65% | 1 | 1 | 0% | 2,916 | 1,482 | -49% | 0 | 0 | — |
case-17 | fail→fail | 8,049 | 6,677 | -17% | 1 | 1 | 0% | 1,231 | 1,361 | +11% | 0 | 0 | — |
case-18 | fail→fail | 1,736 | 2,497 | +44% | 1 | 1 | 0% | 220 | 1,326 | +503% | 0 | 0 | — |
case-19 | pass→fail | 17,404 | 8,892 | -49% | 1 | 1 | 0% | 2,721 | 2,385 | -12% | 0 | 0 | — |
case-20 | pass→fail | 13,336 | 4,430 | -67% | 1 | 1 | 0% | 2,598 | 1,291 | -50% | 0 | 0 | — |
case-21 | pass→fail | 12,785 | 6,281 | -51% | 1 | 1 | 0% | 2,330 | 1,270 | -45% | 0 | 0 | — |
case-22 | pass→fail | 10,688 | 5,284 | -51% | 1 | 1 | 0% | 1,814 | 1,307 | -28% | 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, and 7 counted toward the lift figure. The other 15 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 -33 percentage points is the difference between those two pass rates over the 7 comparable cases. 8 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.