pixazo-media

Generate images, videos, TTS, and music through Pixazo (one API key, 80+ models). Use when Decisions has a Pixazo key or the pixazo MCP is connected.

SKILL.md

GitHub

Pixazo media

Unified image, video, and audio generation via Pixazo.

When to use

  • User wants image, video, TTS, or music and Pixazo is configured (Settings → API Keys → Pixazo).
  • Prefer native tools in the Decisions agent: image_generator, video_generator.
  • In Cursor/Codex, use the pixazo MCP (https://gateway.pixazo.ai/pixazo/mcp) for umbrella routing, or per-model MCPs from pixazo.ai/models/mcp.

Settings

  1. API key: api-console.pixazo.ai/api_keys (sk_live_…)
  2. Enable + Validate in Decisions API Keys (auto-saves).
  3. LLMs tab:
    • Image LLM → Provider Pixazo, pick model (e.g. flux-pro, sdxl-turbo).
    • Video generation → Provider Pixazo, pick model (e.g. seedance-2, veo-3.1).

REST pattern (for scripts)

  1. POST model endpoint or POST https://gateway.pixazo.ai/v1/generate with { "model": "…", "prompt": "…" }
  2. Headers: Ocp-Apim-Subscription-Key and/or Authorization: Bearer <key>
  3. Poll GET https://gateway.pixazo.ai/v2/requests/status/{request_id} until COMPLETED
  4. Download from output.media_url[]

MCP umbrella tools

pixazoGenerateImage, pixazoGenerateVideo, pixazoTextToSpeech, etc. — optional model parameter to target Flux, Seedance, ElevenLabs, etc.

Related

  • fal-ai-media — alternative media MCP when FAL_KEY is set.
  • content-engine / remotion-video-creation — edit and compose finished videos after generation.