Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedWhiteboard Concept Explainer Illustration
v1.0.5Use it to make an image.
npx skillloom-cli install sl-whiteboard-concept-explainer-illustration
Updated Oct 7, 2026

- Install guides
- Claude CodeCodex
- Generates with
- Gpt Image 2
- In the package
7 files · 12.3 KB
- SKILL.md
- references/prompt-framework.md
- references/style-guide.md
- references/visual-system.md
- scripts/generate.mjs
- scripts/image-spec.json
- scripts/validate-image.mjs
Similar skills
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerified
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedMarine Research Poster Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLI
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedSkin Mechanism Visual Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLI
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedAnnotated Wash Diagram Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLI
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedOptical Bench Diagram Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLI
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedRegional Flavor Map Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLI
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedLayered Material Cutaway Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLI
Uses prepaid key: Each run calls an image model, billed to your own prepaid key.ImageVerifiedHand-Rendered Explainer Illustration
Works with: Claude Code, Cursor, OpenCode, Codex, Gemini CLIWhat you get
- One finished concept illustration sized for the selected aspect ratio
- A grid-led composition with a clear visual reading order
- Evidence-grounded visual relationships and restrained annotation
- A prompt tailored to the supplied audience, facts, and learning goal
Good fit
- Teaching a single science or curriculum concept visually
- Publishing a concise explainer in a course or learning resource
- Turning supplied evidence into a clean comparison or process diagram
Not for
- Posters that depend on paragraphs of exact readable text
- Complex datasets, multi-page lessons, or editable diagram source files
Try this prompt
You say
/sl-whiteboard-concept-explainer-illustration Create a 16:9 whiteboard-style explainer for middle-school science students showing how a daytime coastal breeze forms: land warms faster than water; warm air rises over land; cool air moves inland at the surface; air returns aloft toward water and descends over it, forming one closed circulation. Use exactly these labels once: 'Land warms faster', 'Warm air rises', 'Cool air moves inland'. Clean sans-serif; no other text, symbols, numbers, or title.

Then try
Make the same concept suitable for younger students with fewer labels.
Create a comparison version showing daytime and nighttime conditions.
Revise the diagram to use only the supplied source notes and exact labels.
What you provide
- The concept and intended learning outcome
- Audience age or prior knowledge
- Facts or source notes that support the explanation
- Exact labels and preferred image ratio
Requirements
- Node.js 20+
- A SkillLoom generation key (prepaid)
Running cost
Each image is generated with your prepaid SkillLoom generation key (about $0.05 per image at 1K). Subscriptions do not include generation credits.
A typical run costs about $0.05 in image-generation fees, charged to your prepaid key.
Verification
4.0/ 5
Render score
Verified
| Real render check | Verified | 4.0 / 5 | 74s |
|---|
FAQ
Can I include labels?
Yes. Supply exact wording and keep labels short. The illustration should rely primarily on visual relationships, since generated image text can be imperfect.
Will it invent supporting facts?
The prompt should distinguish supplied evidence from interpretation and direct the model not to add unsupported statistics, citations, or causal claims.