muapi-ad-creative

Generate a high-converting ad creative set — hero image, ad copy variations, and platform-optimized crops for Meta, Google Display, and LinkedIn.

Paste into Claude, ChatGPT or Cursor.

Read the source79 lines
muapi-ad-creative/SKILL.md79 lines3.9 KBpushed 100d agoRawView on GitHub
1---
2slug: muapi-ad-creative
3name: muapi-ad-creative
4version: "1.0.0"
5description: Generate a high-converting ad creative set — hero image, ad copy variations, and platform-optimized crops for Meta, Google Display, and LinkedIn.
6acceptLicenseTerms: true
7---
8 
9 
10# Ad Creative Set
11 
12**Generate a high-converting ad creative set — hero image, ad copy variations, and platform-optimized crops for Meta, Google Display, and LinkedIn.**
13 
14## Inputs
15 
16| Name | Type | Required | Default | Description |
17|:---|:---|:---|:---|:---|
18| `product_or_service` | text | yes | — | What is being advertised (e.g. "SaaS project management tool for remote teams"). |
19| `target_audience` | text | yes | — | Who the ad is for (e.g. "startup founders aged 25–40, tech-savvy"). |
20| `campaign_goal` | text | no | awareness | Campaign objective — "awareness", "consideration", or "conversion". |
21| `tone` | text | no | professional, clean, modern | Creative tone and visual style (e.g. "bold and disruptive", "luxury minimal", "friendly and approachable"). |
22| `product_image` | image_url | no | — | Optional product or brand image URL already in the session. |
23 
24 
25## Steps
26 
27This skill has TWO phases. Phase A creates the hero concept for approval; Phase B fans out to platform formats.
28 
29### Phase A — Hero image + Ad copy
30 
31Submit ONE the plan with:
32 
331. **Hero image**`muapi image generate` (model=nano-banana-pro) or `muapi image edit` (model=nano-banana-pro-edit) if `{{product_image}}` is provided:
34 - Aspect ratio: 1:1 (universal starting point).
35 - Prompt must capture: product/service benefit, target audience lifestyle cue, campaign tone.
36 - Style: `{{tone}}, advertising photography, clean background, product focus, ultra detailed, commercial quality`.
37 - Tier: quality.
38 
39After the plan executes, present the hero asset and 3 ad copy variations:
40- **Variation A** — Problem-aware hook: "Tired of X? [Product] fixes that."
41- **Variation B** — Benefit-led: "[Feature] → [Outcome] for [Audience]."
42- **Variation C** — Social proof / urgency: "X teams already use [Product]."
43Each variation includes: Headline (6 words max), Body (20–30 words), CTA button text.
44 
45Ask which copy variation to use for Phase B. Wait for user confirmation.
46 
47### Phase B — Platform crops
48 
49Once the user picks a copy direction, submit a SECOND the plan with parallel crops:
50 
511. `muapi image edit` → 1:1 (Facebook/Instagram feed, 1080×1080)
522. `muapi image edit` → 9:16 (Story/Reels, 1080×1920)
533. `muapi image edit` → 1.91:1 (Facebook feed wide, 1200×628)
544. `muapi image edit` → 1:1 (LinkedIn feed, same as FB)
55 
56For each crop:
57- Prompt: "Reframe for [platform] ad format. Keep product/subject centered and uncropped. Maintain original palette and tone. Leave headroom/footroom for text overlays."
58- All crops run in parallel.
59 
60Return one asset per format with the recommended copy overlay placement for each.
61 
62## Notes
63- If `campaign_goal` is "conversion", emphasize urgency and direct CTA in copy.
64- If `campaign_goal` is "awareness", prioritize visual impact over text density.
65- Reference `product_image` via `$nX.url` syntax in Phase B nodes to ensure consistency.
66- Do NOT auto-confirm Phase B without user picking a copy variation.
67 
68## Trigger Keywords
69 
70`ad creative`, `advertisement`, `facebook ad`, `meta ad`, `google ad`, `linkedin ad`, `paid ad`, `ad banner`, `display ad`
71 
72 
73---
74 
75## Notes for the Executing Agent
76 
77- This recipe is LLM-orchestrated: read each phase, gather any missing inputs from the user, then call `muapi` CLI commands. Use `muapi auth configure` first if `MUAPI_API_KEY` is unset.
78- For model IDs without a CLI alias yet, fall back to the raw endpoint via `curl -X POST https://api.muapi.ai/api/v1/<endpoint> -H "x-api-key: $MUAPI_API_KEY" -H 'content-type: application/json' -d '{...}'` and poll with `muapi predict wait <request_id>`.
79- Substitute `{{input_name}}` placeholders with the user's actual inputs before issuing each call.

Discussion