Video Generate

Generates a video from a text prompt using AI video generation models. Supports both text-to-video and image-to-video (animating a static image).

Command

agent-media video generate --prompt <text> [options]

Inputs

Option	Required	Description
`--prompt`	Yes	Text description of the video to generate
`--in`	No	Input image path or URL for image-to-video
`--duration`	No	Duration in seconds (6, 8, 10, 12, 14, 16, 18, 20; default: 6)
`--resolution`	No	Video resolution (720p, 1080p, 1440p, 2160p; default: 720p)
`--fps`	No	Frame rate (25 or 50; default: 25)
`--audio`	No	Generate audio track
`--out`	No	Output path, filename or directory (default: ./)
`--provider`	No	Provider to use (default: auto-detect)
`--model`	No	Model to use (overrides provider default)

Output

Returns a JSON object with the generated video path:

{
  "ok": true,
  "media_type": "video",
  "action": "video-generate",
  "provider": "fal",
  "output_path": "generated_abc123.mp4",
  "mime": "video/mp4",
  "bytes": 12345678
}

Examples

Generate a video from text:

agent-media video generate --prompt "a cat walking in a garden"

Generate a longer video with higher resolution:

agent-media video generate --prompt "ocean waves crashing on a beach" --duration 10 --resolution 1080p

Animate an existing image (image-to-video):

agent-media video generate --in portrait.png --prompt "the person waves hello and smiles"

Generate with audio:

agent-media video generate --prompt "fireworks exploding in the night sky" --audio

Use a specific provider:

agent-media video generate --prompt "a rocket launching" --provider replicate

Providers

This action requires an external provider:

fal - Requires FAL_API_KEY (uses LTX-2 model)
replicate - Requires REPLICATE_API_TOKEN (uses lightricks/ltx-video model)

The local provider does not support video generation.

video-generate

SKILL.md