Veo Video

Documents

veo-video

Veo Video is an Agent Skill for documents published in the majiayu000/claude-skill-registry repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Generate AI videos with Google Veo via AceDataCloud API. Use when creating videos from text descriptions, animating still images into video, or converting lower-resolution results to full 1080p. Supports Veo 2, Veo 3, and Veo 3.1 models.

View on GitHub ↗

QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

Open your project in Codex.
Copy and paste the prompt below.
Review files and risks before approving any action.

Prompt to paste

Inspect the Agent Skill at https://github.com/majiayu000/claude-skill-registry/blob/HEAD/skills/generation/veo-video-acedatacloud-skills/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/veo-video, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Veo Video Generation

Generate AI videos through AceDataCloud's Google Veo API.

Authentication

export ACEDATACLOUD_API_TOKEN="your-token-here"

Quick Start

curl -X POST https://api.acedata.cloud/veo/videos \
  -H "Authorization: Bearer $ACEDATACLOUD_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "a whale breaching in slow motion at golden hour", "model": "veo-3", "wait": true}'

Models

Model	Duration	Audio	Best For
`veo-2`	5–8s	No	Fast, cost-effective generation
`veo-3`	8s	Yes (native)	Full audiovisual generation
`veo-3.1`	8s	Yes (native)	Latest model, highest quality

Workflows

1. Text-to-Video

POST /veo/videos
{
  "prompt": "cinematic aerial shot of the Northern Lights over Iceland",
  "model": "veo-3",
  "aspect_ratio": "16:9",
  "duration": 8
}

2. Image-to-Video

Animate a still image into video.

POST /veo/videos
{
  "prompt": "the scene gently comes to life with wind and subtle motion",
  "image_url": "https://example.com/landscape.jpg",
  "model": "veo-2",
  "aspect_ratio": "16:9"
}

3. Upscale to 1080p

Convert a generated video to full 1080p resolution.

POST /veo/videos/1080p
{
  "video_url": "https://example.com/generated-video.mp4",
  "model": "veo-3"
}

Parameters

Parameter	Values	Description
`model`	`"veo-2"`, `"veo-3"`, `"veo-3.1"`	Model to use
`aspect_ratio`	`"16:9"`, `"9:16"`	Video aspect ratio
`duration`	`5` – `8`	Duration in seconds
`generate_audio`	`true` / `false`	Enable/disable audio (veo-3, veo-3.1 default to true)
`enhance_prompt`	`true` / `false`	Let the model expand your prompt for better results

Task Polling

POST /veo/tasks
{"task_id": "your-task-id"}

States: pending → succeeded or failed.

MCP Server

pip install mcp-veo

Or hosted: https://veo.mcp.acedata.cloud/mcp

Key tools: veo_generate_video, veo_generate_video_from_image, veo_get_1080p_video

Gotchas

Veo 3 and 3.1 generate native audio — use generate_audio: false to suppress
The 1080p endpoint (/veo/videos/1080p) only upscales previously generated videos
enhance_prompt can significantly improve results but may deviate from literal interpretation
Veo 2 does NOT support audio generation
Duration is capped at 8 seconds for all models
Task states use "succeeded" (not "completed") — check for this value when polling