Generative-Media-Skills

What it is

Generative Media Skills for AI Agents is a multimodal toolset that enables generation, editing, and display of images, videos, and audio for AI agents (Claude Code, Cursor, Gemini CLI) via a schema-driven architecture. It relies on muapi-cli for API access and exposes a collection of core primitives and an expert library, plus an MCP server to expose tools to agents.

How it works

The repository is organized into a Core Primitives section that wraps muapi-cli for raw API access, and an Expert Library with domain-specific skills (e.g., Cinema Director, Nano-Banana, UI Designer, Seedance 2). It provides 41 ready-to-run workflow recipes (recipes) that orchestrate multiple muapi-cli calls into end-to-end pipelines. Executors can pass inputs to skills, which run via muapi CLI calls or direct API endpoints. The MCP server can expose 19 tools to Claude Desktop, Cursor, or any MCP-compatible agent.

Getting started

• Install the muapi CLI (core requirement):

# via npm (recommended — no Python required)
npm install -g muapi-cli

# via pip
pip install muapi-cli

# or run without installing
npx muapi-cli --help

• Configure your API key:

# Interactive setup
muapi auth configure

# Or pass directly
muapi auth configure --api-key "YOUR_MUAPI_KEY"

• Install the Skills:

# Install all skills to your AI agent
npx skills add SamurAIGPT/Generative-Media-Skills --all

# Or install a specific skill
npx skills add SamurAIGPT/Generative-Media-Skills --skill muapi-media-generation

# Install to specific agents
npx skills add SamurAIGPT/Generative-Media-Skills --all -a claude-code -a cursor

• Generate media and use examples:

muapi image generate "a cyberpunk city at night" --model flux-dev

• Run an expert skill example:

bash library/visual/nano-banana/scripts/generate-nano-art.sh \
  --file ./my-source-image.jpg \
  --subject "a glass hummingbird" \
  --style "macro photography" \
  --resolution "2k" \
  --view

Recipe Pack

Forty-one LLM-orchestrated workflow recipes combine multiple muapi-cli calls into end-to-end pipelines (e.g., from a product photo to a cinematic ad). Each skill is described in a SKILL.md the agent reads and follows; you can bring your own consuming agent (Claude Code, Cursor, MCP).

Quick Start notes

• The Core Primitives wrap muapi-cli for raw API access, and there is a dedicated Expert Library with domain-specific capabilities (cinematography, 3D reasoning, UI design, etc.).
• A Model Context Protocol (MCP) server can be started with muapi mcp serve to expose tools to MCP-compatible agents.
• OpenCode usage and the provided OpenCode example show how to clone and run locally with API key configuration.

Traction

• Stars: 3976
• Forks: 447
• Open issues: 4

Licensing

• License: MIT

Getting the latest

• Last push: 2026-08-03