| name | ak:ai-multimodal |
| description | Analyze images/audio/video with Gemini API (better vision than Claude). Generate images (Imagen 4, Nano Banana 2, MiniMax), videos (Veo 3, Hailuo), speech (MiniMax TTS), music (MiniMax). Use for vision analysis, transcription, OCR, design extraction, multimodal AI. |
| user-invocable | true |
| when_to_use | Invoke for Gemini vision, OCR, media generation, or transcription. |
| category | ai-ml |
| keywords | ["vision","image","video","audio","Gemini"] |
| license | MIT |
| allowed-tools | ["Bash","Read","Write","Edit"] |
| argument-hint | [file-path] [prompt] |
AI Multimodal
Process audio, images, videos, and documents with the exact-pinned
@mrgoonie/multix@0.2.0 CLI. Use the npx invocation shown here; do not
install or call a floating global multix.
npx -y -p @mrgoonie/multix@0.2.0 multix --version
Setup
Requires Node.js 20+ and provider keys in process env, project .env, or
~/.multix/.env.
export GEMINI_API_KEY="your-key"
export OPENROUTER_API_KEY="your-key"
export MINIMAX_API_KEY="your-key"
Verify setup:
npx -y -p @mrgoonie/multix@0.2.0 multix check --verbose
The backend pin travels with the skill. Users update it by refreshing the
AgentKit kit; there is no auto-update path inside the skill.
Backend ownership
- Treat the exact-pinned Multix CLI as the runtime contract for covered media
operations; keep this skill focused on orchestration, provider setup, and
examples.
- Report missing keys, FFmpeg, provider access, or
multix check failures as
environment blockers, not kit-loader failures.
- Track missing capability upstream, validate a new exact pin, then refresh the
kit. Do not recreate a parallel AgentKit Python backend unless an accepted ADR
or explicit maintainer decision changes backend ownership.
- When changing the pin, verify stale local-script references, skill metadata,
adapter golden output, and provider setup guidance together.
Quick Start
Analyze media:
npx -y -p @mrgoonie/multix@0.2.0 multix gemini analyze \
--files input.png \
--prompt "Analyze this content" \
--format markdown \
--output analysis.md
Transcribe audio or video:
npx -y -p @mrgoonie/multix@0.2.0 multix gemini transcribe \
--files interview.mp4 \
--prompt "Generate a transcript with timestamps" \
--format markdown \
--output transcript.md
Extract structured data:
