Audit paid-ad copy, images, video, hooks, concepts, format coverage, platform-native fit, message match, creative fatigue, accessibility, and policy across supported platforms.
SkillScore82Maintained
Automatic
◆Marketing & Content pack
Last verified 8/12/2026
What it changes
Passed 18 of 22 test cases with the skill, and 15 of 22 without it.
Without skill68%
With skill82%
Checks it passed that the model alone missed
Assigns a designated owner for the experiment brief.
Categorizes inventory by core concept without counting cosmetic resizes as separate concepts.
Avoids defining universal frequency or CTR thresholds as fixed rules for ad fatigue across platforms.
3 fixed15 passed either way4 still failing0 broke
Pass-rate lift
+14 pp
Tokens
-2%
Agent turns
0%
Response time
-12%
Model gemini-3.6-flashMeasured 8/12/2026v118 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 3
Prompt
Our marketing team needs an audit of our recent Facebook and TikTok ad assets. Can you evaluate our current creative library for compliance, performance, and fatigue, and provide a list of recommended creative refresh briefs complete with hypotheses, target audiences, owners, and key success metrics?
What was checked, without and with the skill
Checked
Without skill
With skill
Avoids defining universal frequency or CTR thresholds as fixed rules for ad fatigue across platforms.
Missed
Passed
Includes supporting performance evidence for each recommended creative refresh brief.
Passed
Passed
Specifies the intended ad format for each recommended creative refresh brief.
Passed
Passed
case-01 · gemini-3.6-flash · 8/12/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md15 linessingle file
Creative Audit
Inventory assets by concept, hook, offer, format, placement, audience, date, and
performance window; cosmetic resizes are not distinct concepts.
Load only active-platform specifications and current policy sources.
Diagnose fatigue only from time-series delivery and response evidence, not age
alone.
Keep creative quality, format compliance, and measured performance separate.
Return prioritized refresh or experiment briefs with hypothesis, audience,
format, owner, success measure, and evidence.
Do not claim a universal refresh cadence, creative count, sound behavior, or
performance threshold. Preserve human review for emotional and cultural judgment.