Analyze one or multiple images with Gemini Flash (vision) via the generateContent REST API. Uses Bun + TypeScript to base64-encode images and print model output. Use for captioning, OCR-like extraction, UI/screenshot analysis, and multi-image comparison.
Install
npx skills add cjkihl/cjkihl --skill gemini-image-analysisnpx skills add cjkihl/cjkihl --skill gemini-image-analysis -a claude-code -gnpx skills add cjkihl/cjkihl --skill gemini-image-analysis -a codex -ggemini skills install https://github.com/cjkihl/cjkihl.git --path packages/skills/gemini-image-analysisgh skill install cjkihl/cjkihl gemini-image-analysisgit clone --depth 1 https://github.com/cjkihl/cjkihl /tmp/cjkihl__cjkihl
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/cjkihl__cjkihl/packages/skills/gemini-image-analysis ~/.claude/skills/gemini-image-analysis # Claude Code
cp -R /tmp/cjkihl__cjkihl/packages/skills/gemini-image-analysis ~/.agents/skills/gemini-image-analysis # Codex, Cursor, Gemini CLI, Copilot, OpenCode