thumbnail-creator
Generate article or newsletter thumbnail candidates using the Gemini API from inside Claude Code. Claude reads article copy, proposes composition concepts, writes image generation prompts incorporating brand specs, calls Gemini to generate the images, evaluates the results via computer vision, and returns ranked candidates with rationale. Use when asked to create thumbnails, generate cover images, or produce visual candidates for an article or newsletter.
npx skills add mohitagw15856/pm-claude-skills --skill thumbnail-creator --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Generates article and newsletter thumbnail candidates by acting as an image-generation agent inside Claude Code. It reads article copy, proposes compositions, writes tailored prompts, calls the Gemini API, evaluates the outputs, and returns ranked results with brief rationale. The output is production-ready thumbnail candidates you can drop directly into your CMS, newsletter tool, or social scheduler.
How it works
- Ingest and analyze the article: Accepts article copy or a URL, fetches title, hook, central claim, key points, statistics, and author name.
- Collect brand specs: Requests brand colours, font style, visual style, and any style references; supports defaults if not provided.
- Propose composition concepts: Generates 3-4 concepts with names, precise layout descriptions, mood/use case, and a rationale tailored to the article; asks user which to generate.
- Write Gemini image generation prompts: Creates detailed prompts per selected composition with exact headline text, style, color hexes, and a negative prompt; keeps prompts under 300 words.
- Check prerequisites and run generation: Verifies GEMINI_API_KEY, existence of generate_image.py, and dependencies; runs generate_image.py for each prompt to produce candidate images.
- Evaluate generated images: Uses computer vision to assess Brand Fit and Text Legibility, with scoring guidance and notes about possible text rendering issues.
- Produce evaluation report: Creates an evaluation_summary.md including scores, rationales, recommended picks per use case, and full prompts used.
- Offer iteration: Provides options to adjust colors, try new compositions, alter headlines, rerun with different Gemini parameters, or generate more variants.
When to use it
Use when asked to create thumbnails, generate cover images, or produce visual candidates for an article or newsletter.
What it can touch
Keys and scripts: relies on the Gemini API (GEMINI_API_KEY) and the generate_image.py script. Uses the filesystem to store outputs under ./thumbnails/[article-slug]/ with candidate image files and evaluation_report.md. The skill expects to find and invoke generate_image.py and may use a prompts.json if batching prompts.
Caveats
The workflow depends on the Gemini Imagen API and the external script; results may vary with API response and image rendering. Text rendering can be imperfect and may require manual overlay adjustments.
# Thumbnail Creator Skill (via Gemini) Generates article and newsletter thumbnail candidates by acting as an image-generation agent inside Claude Code. Instead of switching between tools and prompting Gemini's web UI one image at a time, this skill makes Claude do the full loop: read the copy, propose compositions, write tailored prompts, call the Gemini API, evaluate the outputs, and return ranked results with brief rationale. The output is production-ready thumbnail candidates you can drop directly into your CMS, newsletter tool, or social scheduler. --- ## Prerequisites Both of these must be in place before the skill can generate images: ### 1. Gemini API Key Get a free key from [Google AI Studio](https://aistudio.google.com/app/apikey). Set it as an environment variable: ```bash export GEMINI_API_KEY="your-key-here" ``` To persist it across sessions, add to your shell profile (`~/.zshrc` or `~/.bashrc`): ```bash echo 'export GEMINI_API_KEY="your-key-here"' >> ~/.zshrc source ~/.zshrc ``` Verify it is set: ```bash echo $GEMINI_API_KEY ``` ### 2. generate_image.py Script This script must exist at `./generate_image.py` in the project root. The full template is provided in the Scri
- Prerequisites
- 1. Gemini API Key
- 2. generateimage.py Script
- Required Inputs
- Output Structure
- Phase 1 — Composition Proposals (text, before any API calls)
- Phase 2 — Generated Image Files
- Phase 3 — Evaluation Summary Table
- How Claude Should Execute This Skill
- Step 1 — Ingest and analyse the article
- Step 2 — Collect brand specs
- Step 3 — Propose composition concepts
- Step 4 — Write Gemini image generation prompts
- Step 5 — Check prerequisites and run the generation script
export GEMINI_API_KEY="your-key-here" echo 'export GEMINI_API_KEY="your-key-here"' >> ~/.zshrc source ~/.zshrc echo $GEMINI_API_KEY pip install google-generativeai Pillow requests uv pip install google-generativeai Pillow requests Check API key is set Check script exists ls -la ./generate_image.py Check dependencies
What does the thumbnail-creator skill do?
Generate article or newsletter thumbnail candidates using the Gemini API from inside Claude Code. Claude reads article copy, proposes composition concepts, writes image generation prompts incorporating brand specs, calls Gemini to generate the images, evaluates the results via computer vision, and returns ranked candidates with rationale. Use when asked to create thumbnails, generate cover images, or produce visual candidates for an article or newsletter.
How do I install it?
Run `npx skills add mohitagw15856/pm-claude-skills --skill thumbnail-creator --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From mohitagw15856/pm-claude-skills, a repository with 1,255 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.