/ads-creative

name: ads-creative
description: >
  Cross-platform creative quality audit covering ad copy, video, image, and
  format diversity across all platforms. Detects creative fatigue, evaluates
  platform-native compliance, and provides production priorities. Collects brand
  context and creates branding.md if missing. Uses infographic-v2 for generating
  ad creatives. Use when user says "creative audit", "ad creative", "creative
  fatigue", "ad copy", "ad design", or "creative review".
disable-model-invocation: true

Cross-Platform Creative Quality Audit

Process

1. **Brand context check** — look for `./branding.md`. If not found, run the Brand Info Collection flow below before proceeding. 2. Collect creative assets or performance data from active platforms 3. Read `ads/references/platform-specs.md` for creative specifications 4. Read `ads/references/creative-volume.md` for per-platform volume and refresh requirements 5. Read `ads/references/benchmarks.md` for CTR/engagement benchmarks 6. Read `ads/references/scoring-system.md` for weighted scoring algorithm 7. Evaluate creative quality per platform 8. Run Brand Consistency Audit using branding.md (advisory, not scored) 9. Assess cross-platform creative consistency 10. Generate production priority recommendations 11. **Offer to generate new creatives** using `/infographic-v2` for any flagged gaps

---

Brand Info Collection

Run this at the start if `./branding.md` does not exist in the project root.

**Goal:** Extract the brand's actual styling, copy, and identity — primarily from their website — to personalize the creative audit and enable on-brand creative generation via infographic-v2.

Step 1: Ask for brand name and website

> "Before I audit your creatives, I need your brand context to check consistency. What's your brand name and website URL?"

Ask for the brand name and website URL first. If the user provides a URL, immediately run the website extraction (step 2) before asking any further questions.

Step 2: Website extraction (when URL provided)

**Do not guess or assume anything.** Extract the actual values from the live website.

2a — Fetch the homepage

Use WebFetch on the homepage URL. Extract ALL of the following:

**Visual identity:**

• All colors used prominently: backgrounds, buttons, headers, links, text, borders,

footer. Report exact hex values. Identify which is primary (most prominent brand color, usually on CTAs and headers), secondary, accent, background, and text color.

• Font families: read the actual `font-family` declarations used on headings and body text.

Look in CSS, `<link>` tags for Google Fonts / Typekit / custom font URLs, and inline styles. Report the exact font names (e.g., "Plus Jakarta Sans", "DM Sans", "Geist"), not generic fallbacks like "sans-serif".

• Logo: describe the logo from the header/nav. Note if it's text-based, icon-based, or both.
• Visual style: describe what you observe — minimal with whitespace? Bold with saturated

colors? Dark mode? Gradient-heavy? Illustration-driven? Corporate?

• Button styles: rounded, square, pill-shaped? What color? What text color on buttons?
• Spacing and density: tight and information-dense, or airy and spacious?

**Copy and messaging:**

• Tagline / slogan from the hero section
• Hero headline: the main H1 or hero text
• Value proposition: what they do and why it matters
• Key benefits: any benefit sections, feature highlights, selling points
• Social proof: testimonials, customer logos, metrics, awards, trust badges, star ratings
• CTAs: every call-to-action button text on the page
• Navigation items: main nav labels reveal product/service categories

**Brand voice signals:**

• Tone of the copy: formal or casual? Technical or plain? Playful or serious?
• Person: first person ("We help...") or second person ("You can...")?
• Specific words/phrases: any repeated brand language, trademarked terms

2b — Fetch additional pages (if needed)

If the homepage left gaps, also fetch:

• `/about` or `/about-us` — company story, mission, team, more voice signals
• `/pricing` — pricing model (free trial, freemium, tiers, contact-us)

Only fetch if there are genuine gaps. Don't fetch for the sake of it.

2c — Present extraction to user

Show everything extracted clearly:

Here's what I found on [website]:

  Brand: [name]
  Tagline: "[extracted tagline]"
  Industry: [detected industry]
  Business model: [detected model]

  Colors:
    Primary:    [name] #[hex] — used on [CTAs, headers, etc.]
    Secondary:  [name] #[hex] — used on [subheadings, etc.]
    Accent:     [name] #[hex] — used on [highlights, badges, etc.]
    Background: #[hex]
    Text:       #[hex]

  Fonts:
    Headings: [exact font name]
    Body:     [exact font name]

  Visual style: [description of what you observed]
  Button style: [pill/rounded/square], [color], [hover behavior if visible]

  Voice: [formal/casual], [serious/playful], [technical/simple]
  Hero: "[exact hero headline]"
  Value prop: "[extracted value proposition]"
  CTAs found: "[CTA 1]", "[CTA 2]", "[CTA 3]"
  Social proof: [what's shown — logos, metrics, testimonials]

Does this look right? Anything to correct or add?

The user confirms or corrects. Only after confirmation, proceed with remaining questions.

Step 3: Fill gaps with questions

After website extraction, only ask about things that **could not** be determined from the website. Skip any section where the website gave a clear answer.

**Likely still needed (websites rarely cover these):**

• Target audience: who they sell to (job title/role, age), pain points, buying triggers,

where they spend time online

• Ad-specific guidelines: confirm extracted CTAs, pricing in ads?, required elements

(trust badges, ratings), anything forbidden

• Brand voice refinement (only if ambiguous from the site): words they always/never use

**Do NOT re-ask about colors, fonts, tagline, visual style, or value proposition if the website e