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 |
scripts/repair_patch.md
View on GitHub# repair_patch.py — free official-label pixel repair
`repair_patch.py` copies a real label into a manually verified compatible label band. It makes no model or network call. It does not remove backgrounds, match poses, reconstruct a product, invent lettering or fix a wrong silhouette. Review the actual source and candidate by eye; a passed numerical invariant is not an image-quality verdict.
Write a manifest with distinct new paths and explicit confirmations based on the brand's media rights and your visual inspection:
```json
{
"source_image": "official-product.jpg",
"base_image": "failed-render.png",
"output_image": "repair/candidate.png",
"mask_image": "repair/mask.png",
"evidence_json": "repair/evidence.json",
"source_rights_confirmed": true,
"compatible_pose_reviewed": true,
"source_band": {"left": [120, 120], "right": [360, 360], "top": [200, 205, 200], "bottom": [600, 605, 600]},
"target_band": {"left": [80, 75], "right": [200, 195], "top": [150, 153, 150], "bottom": [350, 353, 350]},
"feather_px": 1.5
}
```
Coordinates are pixels in the full source/base image. `left`/`right` are x coordinates at the top and bottom ends. `top`/`bottom` are y coordinates across the label at left, center and right. The three y guides define each curved boundary; inspect them against the real label edges. The example numbers are illustrative, not a usable brand preset. Use near-front-facing, unoccluded labels with compatible shape and lighting; a severely oblique or differently wrapped label needs another approach.
```bash
python3 repair_patch.py repair-manifest.json
```
The helper validates finite in-bounds, non-crossing geometry and inverse convergence, anti-aliases the source before remapping, feathers within the target boundary and saves a lossless candidate, mask and provenance. It refuses to overwrite the original/source/manifest paths. Every pixel outside the saved mask must remain identical. RGB inputs only; transparent-cutout handling is outside this helper. The agent still checks lettering, edge seams, geometry and lighting at full size.
After a fresh judge passes the full image, upload and save it through the existing GooseWorks tools with its failed creative as parent. Keep both versions and the repair evidence; verify saved media bytes and lineage. This is free local processing plus a free save. Any paid extraction/edit/generation still needs its own applicable quote and approval. An exhausted generation ceiling does not authorize another paid attempt.
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 |