AI-powered content generation platform that turns text into visual media — presentations, greeting cards, comics, infographics, and animated videos — using Google Gemini and Alibaba Dashscope APIs.
概览
AI-powered content generation platform that turns text into visual media — presentations, greeting cards, comics, infographics, and animated videos — using Google Gemini and Alibaba Dashscope APIs. Two implementation stacks share the same config format, style system, and TOML templates: - (mofa/) — Node.js, requires @google/genai, pptxgenjs, sharp - (mofa-cli/) — single binary, zero Node.js dependency - — required for all skills - — optional, for Qwen-Edit image refinement - — for the JS engine - — for video compositing (mofa-video) - (magick) — for comic/infographic stitching (JS engine only) - — for mofa-site Quarto lesson builds - — for mofa-publish deployments Copy the example config and set your API keys: Edit mofa/config.json — API keys use env:VAR_NAME to read from environment variables: For Gemini calls, the Rust CLI uses GEMINI_API_KEY by default. To route Gemini through Google Cloud Vertex AI instead, add a vertex block.
README
以下为原文 README 节选，完整内容请查看原文。
mofa-skills
AI-powered content generation platform that turns text into visual media — presentations, greeting cards, comics, infographics, and animated videos — using Google Gemini and Alibaba Dashscope APIs.
Skills
| Skill | Output | Styles | Description |
|---|---|---|---|
| mofa-slides | .pptx |
14 | Presentation decks with full-bleed AI images and editable text overlays |
| mofa-cards | .png |
7 | Greeting cards in Chinese art styles (ink-wash, guochao, etc.) |
| mofa-comic | .png |
5 | Multi-panel comic strips (xkcd, manga, ligne-claire, etc.) |
| mofa-infographic | .png |
4 | Multi-section infographics stitched vertically |
| mofa-video | .mp4 |
4 | Animated video cards with background music via Gemini Veo |
| mofa-research | text | — | 3-agent deep research pipeline (search → analyze → synthesize) |
| mofa-videolizer | .srt |
— | Subtitle generation from text + audio (Whisper / basic fallback) |
| mofa-workflow | artifacts | — | Multi-agent team pipeline (architect → developer → reviewer → tester) |
| mofa-news | .md |
8 categories | News digest from Google News, Hacker News, Yahoo, Substack, Medium |
| mofa-github | text | — | GitHub integration via gh CLI (issues, PRs, CI, releases, API) |
| mofa-public-apis | text | 40+ categories | Search free public APIs — browse by category, auth, HTTPS, CORS |
| mofa-mcdonalds | text | — | McDonald’s China ordering via MCP — 点餐、领券、积分兑换 |
| mofa-xhs | text | — | Xiaohongshu (小红书) integration — search, read, like, comment, publish |
| mofa-crawler | text | — | Web crawling via Cloudflare Browser Rendering API — full-site extraction with JS rendering |
| mofa-logo | SVG | 8 styles | AI logo generation with Claude Opus 4.6 — minimalist, mascot, emblem, wordmark |
| mofa-fm | .wav |
— | Voice TTS with custom voice cloning via Qwen3-TTS on Apple Silicon |
| mofa-site | static HTML / starter source | Quarto, Astro, Next.js, React | Build lesson sites, scaffold extracted templates, and run a server-backed site studio |
| mofa-publish | live URL | GitHub Pages, Mac Mini | Deploy built static sites and verify the live URL |
DingTalk Wukong Skills (from stvlynn/dingtalk-wukong-skills)
| Skill | Description |
|---|---|
| docx | Advanced Word document creation/editing with tracked changes, comments, formatting preservation |
| xlsx | Excel spreadsheet processing with formulas, data analysis, and visualization |
| pptx | PowerPoint presentation creation/editing, including HTML-to-PPT conversion |
| PDF processing — text/table extraction, creation, merging/splitting, form filling | |
| pdf-convert-to-word | Professional PDF to Word/Markdown conversion |
| dws | DingTalk workspace operations (OA, calendar, docs, todo, chat, approval, attendance, etc.) |
推荐工具
换一个关键词，或者移除筛选条件。
安装
npx skillfish add mofa-org/mofa-skills