gooseworks-ai/goose-skills / skills/marketing/make-static-ads
make-static-ads
VerifiedMake new static image ads, a fresh batch, more ads, a winner's variations, copy variants or a rejected ad fix. Reads the brand, its customers and the ads it already runs, turns real product facts and customer situations into distinct visual ideas, shows a concept board, writes each render prompt, quotes, renders and reviews through GooseWorks. Also handles requested branded header/social graphics. Use for static ad creation rather than goose-ads. Not for video, product photos or launching campaigns.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 14.0 KB |
| references/brand-profile.md | markdown | 4.9 KB |
| references/claims-and-policy.md | markdown | 6.9 KB |
| references/engine-notes.md | markdown | 3.4 KB |
| references/fonts-and-color.md | markdown | 2.5 KB |
| references/formats-and-placements.md | markdown | 2.2 KB |
| references/handoff-format.md | markdown | 4.3 KB |
| references/ideas.md | markdown | 6.5 KB |
| references/levels.md | markdown | 4.2 KB |
| references/prompts.md | markdown | 7.7 KB |
| references/reading-the-brain.md | markdown | 6.0 KB |
| references/review-rubric.md | markdown | 6.6 KB |
| scripts/README.md | markdown | 2.1 KB |
| scripts/place_logo.py | script | 3.6 KB |
| scripts/repair_patch.md | markdown | 2.5 KB |
| scripts/repair_patch.py | script | 9.0 KB |
| scripts/template_sheet.py | script | 3.7 KB |
| skill.meta.json | config | 6.3 KB |
| tests/test-by-hand.md | markdown | 4.9 KB |
| tests/test_place_logo.py | script | 1.9 KB |
| tests/test_repair_patch.py | script | 5.9 KB |
| tests/test_template_sheet.py | script | 1.5 KB |
references/handoff-format.md
View on GitHub# Hand-off format — what the marketer and sibling playbooks receive
## 1. Tags and naming
Write with `gooseworks:ads_creative_update` `patch.tags`. Read the creative's current tags, merge, and write the whole set once; everything lower-case (keys and values, hyphenated):
`made-by:make-static-ads` · `batch:<yyyy-mm-dd>` · `concept:<yyyymmdd>-c<n>` (the board row; the date keeps ids unique across batches) · `style:<slug>` · `angle:<a-nn from the angle bank, else a short slug>` · `type:<angle type, map below>` · `level:<close|balanced|bold>` · `template:<library template id, first 8 characters; or meta-<advertiser-slug>; or own-<creative id, first 8>>` · `mode:<fresh|iterate|rotation|fix>` · `test:<short slug of the test question, e.g. hook-vs-proof>` · `offer-ends:<yyyymmdd>` (promo ads) · `claims:<pass|cuts|facts-only>` · `judge:<pass|needs-review|unjudged>` · `try:<1|2>` · `superseded` · `picked` (only on the marketer's decision)
| Angle family | `type:` |
|---|---|
| Problem → solution | problem_fix |
| How it works / demo | how_it_works / demo |
| Reviews, press, proof | proof |
| Old way vs new | old_vs_new |
| Identity, aspiration, occasion | identity / desire |
| Objection | objection |
| Founder or origin | story |
| Offer | offer |
| Curiosity or news | curiosity / news |
Picks are also written in the batch file, which stays the record of truth. Readable name for the ad account: `brand_product_angle_level_ratio_yyyy-mm-dd`.
## 2. The board at the spend gate (Step 7)
Short, in this order:
1. **Above the cards**, three lines: what was read ("Read: 14 reviews, 20 of ~300 live Meta ads, 15 X posts, the homepage; LinkedIn not checked"); the level and angles, and who chose them; the brand's look in one line.
2. **One card per ad:**
- **The template sheet**, shown to the marketer as the `make-static-ads-find-references` skill (`../../make-static-ads-find-references/SKILL.md`) §5 says: the chosen template (A) and two alternates (B, C) in other styles, each captioned with its source and style.
- **Angle** (yours / new) · **Message** · **Visual idea** (one sentence) · **Style**
- **On-image text**, slot by slot, exactly as it will be drawn
- **Why it's ours:** the product fact or customer words behind it, with the source
- **Quote:** credits for each ratio
3. **Below the cards**, a few lines: the critic's main objection and what changed; cut lines with what would confirm them; open flags; anything the run check marked skipped, and why; this sentence word for word: "If an image fails my check I redo it once, at the same price."
4. **The question**, with the question tool when the host has one (playbook Step 2): "Go ahead with these quotes?" Options: "Yes, make them as quoted" (Recommended), "Swap a template first", "Change or drop an ad"; when no critic ran, add "Tell me which rows to cut". "Swap a template" → one question per ad (up to four per call): "Template for ad <n>?" A (Recommended), B, C.
No real product photos: no quotes; end with "Send me real photos (or I can make them first) and I'll quote these."
Prompts, fonts and palettes stay in the batch file; show them when asked.
## 3. The finished set (Step 9)
One card per ad, with a "Based on" sheet on top, shown to the marketer the same way (`../scripts/template_sheet.py --no-chosen`: the template, then the kept image with its logo as chosen; no shell: the two images side by side):
- Angle · message · visual idea · style · template · level
- On-image text as drawn
- Primary text A and B · link headline · description · CTA · destination
- Claims and platform rules · check result · link
Write the check lines in plain words ("safe to say", "cut 'never oily': nothing on file confirms it", "passed my check on the second try", "please read the headline letter by letter"). Then, in this order (the same content goes into `ads/batches/<YYYY-MM-DD>-<slug>.md`, `reading-the-brain.md` §7):
1. **Cut lines:** `line → ad → why → proof that would restore it`.
2. **Copy variants** (if made), under each parent ad.
3. **Flags and what you should check:** taste calls, regulated-category notes, renders still failing after the retry, anything not checked by eye, and what I checked.
4. **A suggested test:** for example one ad per angle at the same budget. The structure is the marketer's call.
5. **Your pick:** "Tell me which to run; I'll mark them approved for launch."
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |