Image generation
SkillDocs & knowledgeGenerate a standalone image, a visual asset for another artifact, or a complete image-rendered presentation slide. Use when the user asks to generate, illustrate, visualize, make an image for a slide, improve a presentation with generated imagery, or create an image-generated slide. Preserve the explicitly requested output format.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Image generation skill
What this skill tells your AI
The instructions your AI receives, as published by drpedapati/sciclaw in skills/image-generation/SKILL.md and read by ahel’s review.
Identify the final deliverable before generating. Do not switch product types because another tool or skill is available.
Route the request
Standalone image
Generate and attach the requested image.
Visual asset inside another artifact
Generate the illustration, diagram, texture, or analogy for later composition. Keep the parent document or presentation as the final deliverable. Follow its governing skill.
Editable presentation
Read and follow the presentation skills. Build or modify the PowerPoint and keep required content editable. Use generated imagery only as an integrated exhibit when useful.
Complete image-generated slide
Generate the finished 16:9 slide as one image. Use source PowerPoint content when supplied, but do not reconstruct an editable PowerPoint unless asked.
Before generation, read and follow the installed pptx skill. Its presentation communication hierarchy governs the slide: argument and audience-facing content come before layout, and layout comes before aesthetics. Image generation is the rendering method, not a replacement for presentation reasoning.
The generated slide must contain:
- an action title stating the takeaway;
- one coherent argument or explanatory sequence;
- concise audience-facing wording;
- an integrated exhibit or visual analogy that carries meaning;
- a clear implication, intervention, or “so what”;
- readable presentation-scale hierarchy and spacing.
Do not return a wordless conceptual illustration when the user requested a slide. Do not add provenance labels or generation disclaimers unless the user explicitly requests them.
Resolve common wording
| User request | Deliverable |
|---|---|
| “Generate an image” | Standalone image |
| “Generate an image for this slide” | Visual asset for composition |
| “Improve this PowerPoint visually and keep it editable” | Editable PowerPoint |
| “Make an image-generated slide” | Complete worded slide image |
| “Use this PowerPoint's content to make an image-generated slide” | Read the PowerPoint, then generate the complete worded slide image |
When wording remains genuinely ambiguous, infer from the most specific requested noun and recent user correction. Ask only if the choice would materially change the deliverable and cannot be inferred.
Reference behavior
The successful stress-capacity pattern is the qualitative reference: a presentation-ready action title, a three-stage explanation, an integrated capacity/overflow analogy, and a de-escalation takeaway composed as one finished slide image. Reuse the communication hierarchy, not the literal artwork.
Signals
- GitHub stars
- 88
- Forks
- 17
- Last commit
- Jul 2026
Advanced
- Catalog kind
- skill
- Gateway key
image-generation-drpedapati- Source
- github.com/drpedapati/sciclaw