generative-media-skills — Claude Code skill: what it does
generative-media-skills is a Claude Code skill that turns creative briefs into muapi.ai calls for directed video, product ads, logos, UI mockups and social covers.
MIT 212 installs ★ 35 updated 2026-10-21
npx skills add podo/design-agent-skills --skill generative-media-skills -a claude-codeWhat generative-media-skills does
Thirteen skills, each one turning a creative brief into calls to muapi.ai, the platform that fronts Midjourney v7, Flux Kontext, Kling 3.0, Veo3, Seedance 2.0 and Suno. A single skill carries prompt templates, parameter choices and the model pick for one output type: a shot list for video work, a product ad assembled from a person photo, a product photo and a script, a logo, a UI mockup.
The set covers muapi-cinema-director and muapi-seedance-2 for directed video, muapi-nano-banana, muapi-logo-creator and muapi-ui-design for stills, logos and mockups, and the rest for short-form, social posts and clipping. Your agent picks the matching one from the task. The upstream repo keeps further recipes under library/<group>/<name>, installed by path: npx skills add SamurAIGPT/Generative-Media-Skills/library/motion/music-video.
Set this up first: a muapi.ai account, the key in MUAPI_API_KEY, the CLI through npm install -g muapi-cli and then muapi auth configure. Nothing runs without the account. Generation burns muapi credits and sends your prompts and uploads to a third-party service. Workflows built around fal.ai belong to fal-ai-skills instead, and plain retouching of existing stills is outside what these recipes do.
What it helps with
- turn a product photo and a script into a video ad
- draft a cinematic shot list for a video model
- generate a logo or a UI mockup from a brief
- cut a long video into short clips for social
- produce a YouTube thumbnail or a Short
Example prompts for the agent
- Build a product video ad from
product.pngand the script inscript.mdwithmuapi-product-video-ad-maker. - Draft the shot list and generate the clip using
muapi-cinema-director. - Generate a logo from
brief.mdwithmuapi-logo-creator.
How to install generative-media-skills
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill generative-media-skills -a claude-code2Manually: put the files in place
~/.claude/skills/generative-media-skills/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/generative-media-skills/
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill generative-media-skills -a codex2Manually: put the files in place
.agents/skills/generative-media-skills/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill generative-media-skills -a cursor2Manually: put the files in place
.cursor/skills/generative-media-skills/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install podo/design-agent-skills skills/generative-media-skills2Manually: put the files in place
.github/skills/generative-media-skills/SKILL.md1Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill generative-media-skills -a gemini-cli2Platform CLI
gemini skills install https://github.com/podo/design-agent-skills --path skills/generative-media-skillsManually: put the files in place
.devin/skills/generative-media-skills/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add podo/design-agent-skills --skill generative-media-skills -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about generative-media-skills
How do I install generative-media-skills in Claude Code?
Run npx skills add podo/design-agent-skills --skill generative-media-skills -a claude-code. The skill lands at ~/.claude/skills/generative-media-skills/SKILL.md.
What does generative-media-skills do?
It packages skills for media generation. Each recipe spells out the prompt pattern, parameters and model for one output, then calls muapi.ai to produce the image or clip.
What do I need before the first run?
A muapi.ai account with the key set in MUAPI_API_KEY, plus the CLI: npm install -g muapi-cli, followed by muapi auth configure. Every generation uses muapi credits.
Source text
SKILL.md · MIT
--- name: generative-media-skills description: | Skills for AI media generation through the muapi.ai platform: cinematic video direction, image generation with Nano Banana, Seedance 2.0 video, UGC and product video ads, logos, UI mockups, YouTube thumbnails and shorts, social posts and AI clipping. Routes to models such as Midjourney v7, Flux Kontext, Kling 3.0 and Veo3. Requires a muapi account and the muapi CLI. triggers: - "generative media" - "Flux Kontext" - "Kling 3.0" - "Veo3" - "SamurAIGPT" das: type: package category: creative-3d upstream: "https://github.com/SamurAIGPT/Generative-Media-Skills" version: latest install: false --- # generative-media-skills > Catalogue stub — full package: [SamurAIGPT/Generative-Media-Skills](https://github.com/SamurAIGPT/Generative-Media-Skills) <!-- das:install:start — generated from stub.yaml by `npm run generate`; edit stub.yaml, not this block --> ## Decision tree 1. **Is the full skill already installed?** This package installs 13 skills (`muapi-ad-creative`, `muapi-ai-clipping`, `muapi-cinema-director`, `muapi-instagram-post`, `muapi-logo-creator`, `muapi-nano-banana`, `muapi-product-video-ad-maker`, `muapi-seedance-2`, `muapi-social-media-video`, `muapi-ugc-video-factory`, `muapi-ui-design`, `muapi-youtube-shorts`, `muapi-youtube-thumbnail`); checking the first. - Global: `f=~/.agents/skills/muapi-ad-creative/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"` - Project: `f=.agents/skills/muapi-ad-creative/SKILL.md; [ -f "$f" ] && ! grep -q "^das:" "$f" && echo "installed" || echo "pointer"` - `installed` → the skill is installed as `muapi-ad-creative`, invoke it and proceed - `pointer` → go to step 2 2. **Detect scope, then install:** ```bash [ -e ~/.agents/skills/generative-media-skills ] && echo "global" || echo "project" ``` **Global:** ```bash npx skills add SamurAIGPT/Generative-Media-Skills --skill muapi-ad-creative muapi-ai-clipping muapi-cinema-director muapi-instagram-post muapi-logo-creator muapi-nano-banana muapi-product-video-ad-maker muapi-seedance-2 muapi-social-media-video muapi-ugc-video-factory muapi-ui-design muapi-youtube-shorts muapi-youtube-thumbnail -g -y ``` **Project:** ```bash npx skills add SamurAIGPT/Generative-Media-Skills --skill muapi-ad-creative muapi-ai-clipping muapi-cinema-director muapi-instagram-post muapi-logo-creator muapi-nano-banana muapi-product-video-ad-maker muapi-seedance-2 muapi-social-media-video muapi-ugc-video-factory muapi-ui-design muapi-youtube-shorts muapi-youtube-thumbnail -y ``` > **Claude Code:** send a command as a chat message starting with `!` to run it without leaving the conversation. **Heads-up** - **Requires:** a muapi.ai account and API key (MUAPI_API_KEY); every skill calls the muapi service and generation consumes muapi credits; the muapi CLI (npm install -g muapi-cli, then muapi auth configure) and curl; generated prompts and uploaded media are sent to the third-party muapi.ai service - **License:** MIT <!-- das:install:end --> ## Invoke after install Every skill is named `muapi-<recipe>`; the agent picks one from the task: - `muapi-cinema-director`, `muapi-seedance-2` — directed cinematic video - `muapi-product-video-ad-maker`, `muapi-ugc-video-factory`, `muapi-ad-creative` — product and UGC ads - `muapi-nano-banana`, `muapi-logo-creator`, `muapi-ui-design` — image generation, logos, UI mockups - `muapi-youtube-thumbnail`, `muapi-youtube-shorts`, `muapi-social-media-video`, `muapi-instagram-post`, `muapi-ai-clipping` — social and short-form content Before first use: install the muapi CLI and run `muapi auth configure` with your muapi.ai API key (see the heads-up above). ## What it does Structured workflow recipes that turn a creative brief into calls to the muapi.ai platform, which fronts many image, video and audio models (Midjourney v7, Flux Kontext, Kling 3.0, Veo3, Seedance 2.0, Suno and others). Each skill encapsulates prompt patterns, parameter choices and model selection for one output type, for example a cinematic shot list for video models or a UGC ad built from a person photo, a product photo and a script. The skills CLI installs the packaged set listed above. The repo holds many more recipes under `library/<group>/<name>`; install one by path, for example `npx skills add SamurAIGPT/Generative-Media-Skills/library/motion/music-video`. Everything runs against a third-party paid service: prompts and uploaded media leave your machine and generation consumes muapi credits. ## When NOT to use - Without a muapi account — the skills require muapi CLI authentication - fal.ai model workflows specifically — use fal-ai-skills instead - Static image editing without AI generation
© podo · MIT · podo/design-agent-skills