S30 · Pipeline skill
Prepare visual material
Collect permitted assets, generate visual candidates and check facts before assembling presentation material.
Permitted asset preparation + visual generation + human fact review and export
- Pipeline
- 4 steps
- Human review
- 1
- Components
- 8
01Pipeline
Base pipeline
Ordered steps. Each consumes the previous output; review gates hold until a person accepts.
- Starts from
- Reviewed text, reference assets and usage permission
- Delivers
- Presentation material with searchable text
Review contract
People decide at review gates; components supply data, drafts or review surfaces. Unaccepted outputs do not advance.
Integration note
Person-image generation is an optional path requiring authorized references; image checks cover availability and size, while humans verify facts.
- 01
Prepare reference assets
- Input
- Reviewed text, reference assets and usage permission
- Output
- Normalized images and use records
- Entity logo discovery and quality reviewEntity website matching + icon normalization and fallback marks
- Seeded avatar generation and normalized custom avatarsSeeded graphic rendering + uploaded image normalization
- 02
Create visual candidates
- Input
- Normalized images and use records
- Output
- Candidate images with text and source records
- AI destination/story image generationReviewed visual prompts + generated images and artifact manifests
- Batch sports image generation and cutoutsReference-based pose generation + cutouts and asset catalogs
- Branded social-card image compositionExisting images with localized text + channel-sized outputs
- 03
Check visual content
Human review- Input
- Candidate images with text and source records
- Output
- Human-confirmed visuals and revisions
Decided byAssigned reviewer or requesting user
- Remote image quality auditingRemote image availability checks + size and quality filtering
- 04
Assemble and export
- Input
- Human-confirmed visuals and revisions
- Output
- Presentation material with searchable text
- Branded pitch deck generation and exportReviewed narratives to slide decks + export previews
- Searchable presentation packagingRaster slide appearance preservation + searchable text layers
02Industry versions
Industry versions
Education Transferable use Turn reviewed course notes into visual cards and slides, with instructors checking that illustrations explain concepts correctly.
Adaptation Generated images cannot serve as experimental results or historical evidence; replace original character assets and templates.
- 01
Prepare course references
- Input
- Course notes, authorized visual references and presentation requirements
- Output
- Normalized course images and use records
- Entity logo discovery and quality reviewEntity website matching + icon normalization and fallback marks
- Seeded avatar generation and normalized custom avatarsSeeded graphic rendering + uploaded image normalization
- 02
Create course-card candidates
- Input
- Normalized course images and use records
- Output
- Candidate cards and explanatory text
- AI destination/story image generationReviewed visual prompts + generated images and artifact manifests
- Batch sports image generation and cutoutsReference-based pose generation + cutouts and asset catalogs
- Branded social-card image compositionExisting images with localized text + channel-sized outputs
- 03
Instructor checks concepts
Human review- Input
- Candidate cards and explanatory text
- Output
- Confirmed visuals and revision list
Decided byAssigned reviewer or requesting user
- Remote image quality auditingRemote image availability checks + size and quality filtering
- 04
Export searchable courseware
- Input
- Confirmed visuals and revision list
- Output
- Searchable slides and previews
- Branded pitch deck generation and exportReviewed narratives to slide decks + export previews
- Searchable presentation packagingRaster slide appearance preservation + searchable text layers
Sports Transferable use Turn authorized athlete references and explanations into training cards, then export teaching material after a coach checks the illustrated actions.
Adaptation Generated images are illustrative, not performance evidence; do not infer injury or results from cards.
- 01
Prepare authorized image references
- Input
- Authorized athlete references, training explanations and templates
- Output
- Normalized reference images and use records
- Entity logo discovery and quality reviewEntity website matching + icon normalization and fallback marks
- Seeded avatar generation and normalized custom avatarsSeeded graphic rendering + uploaded image normalization
- 02
Create training-card candidates
- Input
- Normalized reference images and use records
- Output
- Action illustrations and explanations
- AI destination/story image generationReviewed visual prompts + generated images and artifact manifests
- Batch sports image generation and cutoutsReference-based pose generation + cutouts and asset catalogs
- Branded social-card image compositionExisting images with localized text + channel-sized outputs
- 03
Coach checks illustrations
Human review- Input
- Action illustrations and explanations
- Output
- Confirmed cards and revisions
Decided byAssigned reviewer or requesting user
- Remote image quality auditingRemote image availability checks + size and quality filtering
- 04
Export training presentations
- Input
- Confirmed cards and revisions
- Output
- Searchable training slides
- Branded pitch deck generation and exportReviewed narratives to slide decks + export previews
- Searchable presentation packagingRaster slide appearance preservation + searchable text layers
03Parameters
Parameters
From the pack’s params.schema.json. Choose values to build a synthetic request.
skills/_pipelines/prepare-visual-material/params.schema.json
{
"synthetic": true,
"run_id": "synthetic-example",
"locale": "en",
"industry": "base",
"adapter_mode": "local",
"records": [
{
"id": "synthetic-record-1",
"synthetic": true,
"text": "Synthetic observation A: a draft needs review."
},
{
"id": "synthetic-record-2",
"synthetic": true,
"text": "Synthetic observation B: retain the unresolved question."
}
],
"decisions": {}
} 04Member components
Member components
-
Entity logo discovery and quality review
Entity website matching + icon normalization and fallback marks
Working 1 pipeline -
Seeded avatar generation and normalized custom avatars
Seeded graphic rendering + uploaded image normalization
Working 1 pipeline -
AI destination/story image generation
Reviewed visual prompts + generated images and artifact manifests
Working 1 pipeline -
Batch sports image generation and cutouts
Reference-based pose generation + cutouts and asset catalogs
Working 1 pipeline -
Branded social-card image composition
Existing images with localized text + channel-sized outputs
Working 1 pipeline -
Remote image quality auditing
Remote image availability checks + size and quality filtering
Working 1 pipeline -
Branded pitch deck generation and export
Reviewed narratives to slide decks + export previews
Working 1 pipeline -
Searchable presentation packaging
Raster slide appearance preservation + searchable text layers
Working 1 pipeline
05Skills pack
Skills pack
Runnable reference on synthetic fixtures with tests. Domain contracts, services and approvals still need integration.
prepare-visual-material-1.0.0.zip
Contains SKILL.md, pipeline.json, params.schema.json, runner, tests, fixtures, integration checklist and sources.
Synthetic workflow reference; service integration pending
Download · 20KBSHA-256 c67376714dd7b68db960c76ed9005fc952844269c0f84cc119f0348b3f0b50f7