agentleFS
TuminhaTuminha/dental-ai-skills/dental-image-generator/SKILL.md

dental-image-generator

Skill7 starsChanged 47 days agoMITView on GitHub

Use when the user asks to generate dental clinical illustrations, patient infographics, branded dental visuals, surgical diagrams, anatomy comparisons, post-op visual instructions, or social media graphics using the included OpenAI image-generation script.

  • Reads credentials

The file

Files
SKILL.md

Dental Image Generator — AI Clinical Illustration Skill

Skill protocol version: 2026.09.30

Identity

You generate clinical illustrations, patient infographics, and branded dental visuals using AI image generation. You understand dental anatomy, clinical procedures, and how to communicate visually to different audiences (clinicians vs patients).

Setup

This skill uses OpenAI's Images API with the model gpt-image-2.5-sunburst by default. OpenAI's image-generation guide names it as the model for work where editing precision matters, and the brand-asset option below uses the editing route. It is a paid API: OpenAI bills each image by output tokens, and the current rates are on its pricing page. The model id, the size rules and the quality values were checked live on 2026-09-30 against the API and the guide at https://developers.openai.com/api/docs/guides/image-generation.

Get Your API Key (2 minutes)

  1. Go to https://platform.openai.com/api-keys
  2. Sign in and click "Create new secret key"
  3. Copy the key. OpenAI may ask you to complete API Organization Verification in the developer console before the key can use the GPT Image models (https://developers.openai.com/api/docs/guides/image-generation, read 2026-09-30).
# Add to your shell profile (~/.zshrc, ~/.bashrc)
export OPENAI_API_KEY="your-api-key-here"

Install Dependencies

None. The script uses only the Python standard library (Python 3.10 or newer).


Usage

Command Line

python3 scripts/generate_dental_image.py \
  --prompt "Your description here" \
  --style clinical \
  --output output.png

--help and --dry-run work without a key. A real run without OPENAI_API_KEY prints one sentence and exits with code 2. Nothing is printed that contains the key.

Options

OptionDefaultNotes
--prompt, -prequiredWhat to illustrate
--style, -sclinicalclinical, patient-friendly or infographic; each preset is a prefix on the prompt
--output, -ooutput.png.png, .jpg or .webp picks the output format
--model, -mgpt-image-2.5-sunburstAny OpenAI image model id, for example gpt-image-2.5-flare or gpt-image-2
--size1024x10241536x1024 (landscape) and 1024x1536 (portrait) are the other standard sizes. Custom WIDTHxHEIGHT: both edges multiples of 16, aspect between 1:3 and 3:1, 655,360 to 8,294,400 pixels in total
--quality, -qmediumlow, medium, high, xhigh, max or auto; low is the cheapest draft setting
--brand-asset, -bnoneA clinic asset (png, jpg or webp) sent as an input image, see below
--dry-runoffPrint the request as JSON and exit without calling the API

Style Presets

StyleLookBest For
clinicalMedical textbook — clean lines, anatomical labels, neutral colorsSurgical diagrams, anatomical illustrations, clinical comparisons
patient-friendlySoft, calming palette — simple shapes, reassuring, no scary imageryPost-op handouts, waiting room posters, patient education
infographicModern layout — numbered steps, icons, clear hierarchySocial media, educational carousels, quick-reference guides

Brand Reference

Give the script your clinic's logo, brochure or business card and the model matches its colours and design style:

python3 scripts/generate_dental_image.py \
  --prompt "Post-operative implant care instructions" \
  --style patient-friendly \
  --brand-asset /path/to/my-clinic-logo.png \
  --output branded-instructions.png

How it works: the asset goes to the Images API edits endpoint as an input image (png, jpg or webp, under 15 MB: it travels inline as base64 and the API caps that at 20,971,520 characters), with a note in the prompt to reuse its palette, typography feel and overall style without copying the asset itself. There is no separate text-analysis step. Input images add input tokens to the bill.


Prompt Cookbook

Tested prompts that produce reliable results.

Clinical Illustrations

"Create a clinical illustration showing the stages of dental implant placement
in cross-section view. Clean, professional medical illustration style with
labeled anatomical structures."

"Illustrate a comparison between healthy periodontium and stage III
periodontitis, showing bone loss, pocket depth, and inflammation.
Medical textbook style."

"Show a step-by-step sinus lift procedure (lateral window approach) in
4 panels. Label the sinus membrane, bone graft material, and implant site."

"Cross-section illustration of a tooth with apical periodontitis showing
the abscess, periapical radiolucency, and path of infection."

Patient Education

"Design a patient-friendly infographic about post-extraction care
instructions. Use simple icons, numbered steps, and a calming blue
color palette."

"Create a visual guide showing proper brushing technique in 4 steps.
Friendly, cartoon-style, suitable for a dental office waiting room poster."

"Illustrate the difference between gingivitis and periodontitis in a
simple side-by-side comparison. Patient-friendly — no scary imagery."

Social Media / Infographics

"Design a modern infographic titled '5 Foods That Strengthen Your Teeth'.
Clean layout, food icons, brief text for each item. Instagram-ready."

"Create a visual showing the timeline of dental implant healing: surgery →
osseointegration → abutment → crown. Modern, minimal style."

"Design a myth vs fact infographic about dental X-ray safety.
Two columns, checkmarks and X marks, professional but approachable."

Prompt Tips

  1. Be anatomically specific — "cross-section view," "sagittal plane," "buccal aspect"
  2. Specify the audience — "medical textbook" vs "patient handout" produces very different results
  3. Name the structures — "label the alveolar bone, PDL, cementum, and gingiva"
  4. Set the color mood — "calming blue palette" for patients, "high contrast" for clinical
  5. Use panel layouts — "show in 4 panels" or "step-by-step from left to right"
  6. Iterate — first generation not perfect? Add detail and regenerate

Important

Always review AI-generated medical illustrations for anatomical accuracy before clinical use.

Methodology Review Date

Last review: 2026-09-30

  • 2026-09-30: Moved from Google Gemini to OpenAI's Images API. The old script pinned gemini-2.0-flash-exp; Google's deprecations page (https://ai.google.dev/gemini-api/docs/deprecations, read 2026-09-30) lists gemini-2.0-flash as shut down on 1 June 2026 and has no row for the -exp id, so the script could no longer run. The new script is standard library only, defaults to gpt-image-2.5-sunburst, adds --model, --size, --quality and --dry-run, and sends a brand asset as an input image to the edits endpoint instead of running a text analysis first. Model id, size rules and quality values checked live against the API and OpenAI's image-generation guide on that date. The free-tier rate claim was removed; the API is paid.

Re-review this skill when OpenAI's image models page retires gpt-image-2.5-sunburst or lists a newer default, or when the Images API changes its size or quality rules.


Part of Dental AI Skills by Francisco Teixeira Barbosa

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.