What problem does it solve?
Educational content often accepts first-pass visuals that are generic, mislabeled, or inaccessible, leaving learners confused and creating orphan assets that never get embedded into lessons. This Skill replaces that brittle process with a principled, multi-turn generation workflow that produces pedagogically effective, production-ready images and ensures each visual is validated, unique, and immediately integrated into the course.
Core Features & Use Cases
- Multi-turn reasoning partnership with Gemini using creative briefs (story, intent, visual metaphor) rather than pixel specs to activate reasoning and improve aesthetic and pedagogical decisions.
- Six explicit quality gates (spelling accuracy, layout precision, color accuracy, typography hierarchy, teaching effectiveness, uniqueness) to ensure production readiness before download.
- Browser-based Playwright MCP workflow guidance for interacting with gemini.google.com, downloading full-size images, and copying assets into robolearn-interface static directories.
- Batch mode with token conservation and autonomous execution for large visual sets, including checkpoint files for interruption recovery and a reflection artifact for continuous improvement.
- Immediate embedding workflow that inserts images into the correct lesson at the right pedagogical insertion point to avoid orphan assets.
- Use cases: batch-generate chapter visuals for a Docusaurus course, iterate on failed typography or color gates for accessibility, and produce unique, curriculum-aligned infographics.
Quick Start
Paste a concise creative brief into Gemini, run the six quality gates, download a passing image, and embed it into the target lesson file.
