gabelul
UserImage generation skill for Claude Code, Codex CLI, OpenCode, Cursor and 50+ AI coding agents. Two backends (OpenAI API or codex CLI signed in to ChatGPT subscription), three modes (generate, edit, compose up to 16 references), six production recipes for gpt-image-2.
Categories
Indexed Skills (2)
pixeltamer
Generate, edit, and compose images with gpt-image-2. Two backends — OpenAI API key or codex CLI (uses your ChatGPT subscription, no key needed). Three modes — one-shot generate, multi-image batch with verification, and multi-reference composition (up to 16 inputs blended into one). Use when the user asks to make an image, generate a poster, draw a mockup, design an icon, create an ad creative, build an infographic, blend reference images, edit or inpaint an existing image, or anything else that ends in a PNG.
pixelslop
Browser-first design quality review and fix. Scans real pages with Playwright, scores 5 measured pillars, detects AI slop patterns, and runs a design-director pass for subjective judgment findings. Evaluates against 8 built-in personas plus project-specific ones generated from your audience, tracks score trends across runs, and fixes issues toward your design tokens with checkpoint-based rollback. Exhaustive by default (--fast for a quick pass).
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.