muapi-media

Discover MuAPI image models and run one confirmed asynchronous image generation request.

3.4k|487|Updated Jul 25, 2025
One-click install
npx skills add https://github.com/davepoon/buildwithclaude --skill muapi-media
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: muapi-media
Source: https://github.com/davepoon/buildwithclaude/tree/main/plugins/all-skills/skills/muapi-media
Command: npx skills add https://github.com/davepoon/buildwithclaude --skill muapi-media

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

It removes the risk of calling stale or hard-coded MuAPI model endpoints by querying the live model catalog first, and it prevents accidental billable API calls by requiring explicit confirmation before any paid image generation request.

Core Features & Use Cases

  • Live Model Discovery: Query the current MuAPI catalog for text-to-image and image-to-image models with exact versioned endpoints.
  • Confirmed Single Generation: Submit exactly one image generation POST with a mandatory --confirm-paid flag, then poll the prediction result endpoint with a bounded budget.
  • Safe Output Download: Optionally download the first completed artifact over HTTPS without forwarding the API key to the output host.
  • Use Case: An agent needs a product photo via the current flux-dev model; it lists image models, prepares a JSON params file with a prompt, and runs one confirmed generation that polls until completion.

Quick Start

Set the MUAPI_API_KEY environment variable, then ask the agent to list current MuAPI image models and generate one image from a JSON params file with paid confirmation.

Frequently Asked Questions about muapi-media

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an image with the MuAPI API from Python?

Run the bundled muapi_media.py CLI with the generate subcommand, passing a model name from the live catalog, a JSON params file containing a prompt, and the --confirm-paid flag. The script submits one POST and polls the prediction result endpoint until completion.

How do I find current MuAPI image model endpoints?

Run the models subcommand with --category image and an optional --query filter. It queries the live catalog at /api/v1/models and returns each model's name, category, and already-versioned endpoint path.

Does the MuAPI skill support video or audio generation?

No. It only covers text-to-image and image-to-image categories from the MuAPI catalog. Video, audio, 3D, chat, and model-training requests are explicitly out of scope.

Why does MuAPI generation fail without the confirm flag?

The CLI requires --confirm-paid because generation requests may incur charges. Without it, the script raises an error before sending any POST, ensuring no billable request happens without explicit user approval.

What happens if a MuAPI prediction never finishes polling?

The CLI polls with GET requests only and stops after --max-polls attempts (default 60). If the prediction has not completed by then, it raises an error reporting the request ID instead of retrying the generation POST.