Route and coordinate an end-to-end design-award workflow across winner research, evidence-based evaluation, award matching, entry-text preparation, and final submission checking.
SkillScore83Maintained
Automatic
◆Design & UX Review pack
Last verified 8/17/2026
What it changes
Passed 19 of 23 test cases with the skill, and 9 of 23 without it.
Without skill38%
With skill83%
Checks it passed that the model alone missed
Selects design-award-search as the route.
Refuses to estimate a winning probability.
Selects design-submission-check as the route.
and 7 more in the full report
11 fixed9 passed either way4 still failing0 broke
Pass-rate lift
+46 pp
Tokens
-1%
Agent turns
0%
Response time
-29%
Model gemini-3.6-flashMeasured 8/17/2026v220 of 24 cases passed with skill · 23 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 10
Prompt
Our team built 'AuraGlow', an ambient lighting system, and we only want to see previous award winners in the smart home lighting category from the last three years to analyze design trends. Which single specialist route handles this request?
What was checked, without and with the skill
Checked
Without skill
With skill
Selects design-award-search as the route.
Missed
Passed
case-12 · gemini-3.6-flash · 8/17/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md50 lines3 filesincludes scripts
Design Award Pipeline
Coordinate the specialist Design Judge skills without duplicating their rules.
Route the request
Choose the smallest sufficient route:
User goal
Route
Find comparable winners
design-award-search
Diagnose or score the design
design-evaluation
Select an award, route, or category
design-award-match
Prepare entry fields and copy
design-information-prep
Audit a concrete submission package
design-submission-check
Complete journey
evaluation or match → information prep → submission check; add search only when precedents are needed
If the user has not provided enough information to select a route, ask at most one short question. Otherwise proceed with explicit assumptions.
Maintain the handoff
Create or update a compact handoff record using references/handoff-schema.json. Keep these concepts separate:
user-supplied facts;
evidence-backed findings;
model inferences;
missing facts and live-rule checks;
decisions already approved by the user;
the next recommended specialist skill.
Do not copy long specialist outputs into the handoff. Store identifiers, decisions, blockers, source links, and artifact paths.
When a handoff is written to JSON, run python scripts/validate_handoff.py <handoff.json> before treating it as resumable. Use examples/handoff.example.json as a minimal valid shape, not as project evidence.
Coordinate stages
State the selected route and why it is sufficient.
Invoke or follow the relevant specialist Skill exactly.
Preserve its uncertainty labels, evidence confidence, and blockers.
Stop for user approval when the target award, maturity track, or another consequential choice changes downstream work.
Extract evidence-grounded project facts from user-provided design attachments, identify missing information, and prepare the exact written fields required by supported design-award entry forms. Use when a user asks to prepare, draft, adapt, translate, or validate application text for iF, iF Student, Red Dot Product Design, IDEA, DIA, K-Design, GOOD DESIGN AWARD Japan, Core77, James Dyson, or EPDA. Also use to build a reusable project dossier from briefs, decks, reports, manuals, patents, researc