# vidu-video

vidu-video is a Claude Code skill for planning Vidu Open Platform API video generation. It covers prompts, image references, start/end frames, and Q2 multi-frame jobs.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 136 installs · ★ 252 repo stars
- Author: calesthio
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/calesthio/generative-media-skills/tree/HEAD/skills/providers/video-generation/vidu-video/SKILL.md
- Updated: 2026-07-14
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/vidu-video

## What it does

vidu-video designs requests and integrations for the Vidu Open Platform API. It selects an endpoint, model, and mode for text-to-video, image-to-video, start/end-frame, reference-to-video, or multi-frame work, then accounts for media limits, aspect ratios, duration, audio, and pricing.

The skill defaults to a dry-run plan. Paid generation, private-media uploads, task cancellation, and result downloads wait for approval of the provider, endpoint, model, inputs, request digest, maximum charge, storage location, and disclosure plan. The API uses asynchronous tasks and an API key; consumer credits and plans from `vidu.com` are a separate surface.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add calesthio/generative-media-skills --skill vidu-video -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/vidu-video/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/vidu-video/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add calesthio/generative-media-skills --skill vidu-video -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/vidu-video/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add calesthio/generative-media-skills --skill vidu-video -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/vidu-video/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install calesthio/generative-media-skills skills/providers/video-generation/vidu-video
```

2. **Manually: put the files in place**
- File: `.github/skills/vidu-video/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add calesthio/generative-media-skills --skill vidu-video -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/calesthio/generative-media-skills --path skills/providers/video-generation/vidu-video
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/vidu-video/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add calesthio/generative-media-skills --skill vidu-video -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/calesthio/generative-media-skills/tree/HEAD/skills/providers/video-generation/vidu-video/SKILL.md

## When to use it
- choose a Vidu endpoint and model for a video request
- build a request from a prompt, image, or ordered frame pair
- keep named subjects consistent across reference-to-video output
- plan an asynchronous task with an approved charge and storage target

## Example prompts
- Plan a dry-run text-to-video request for the Vidu Open Platform using `viduq3-turbo`.
- Design an image-to-video integration for one reference image and check the file constraints before submission.
- Compare `start-end2video` options for two ordered frames, a chosen duration, and a maximum charge.

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: dedf4eb695c6270f.

BEGIN THIRD-PARTY FILE dedf4eb695c6270f
```text
---
name: vidu-video
description: Plan and integrate ShengShu Vidu Open Platform video generation with current model/mode selection, reference consistency, pricing, exact approval, async task handling, and safe artifact custody. Use for Vidu API text-to-video, image-to-video, start/end-frame, multi-subject reference, media-reference, or Q2 multi-frame work; do not conflate the API with vidu.com consumer plans or Vidu-S1 streaming digital humans.
---

# Vidu video

Use this skill to design a production request or integration for the **Vidu Open Platform API**. Default to dry-run planning. Never submit a paid generation, upload private media, cancel a task, or download a result until the operator has approved the exact provider, endpoint, model, inputs, request digest, maximum charge, storage destination, and disclosure plan.

This document was verified against first-party public documentation on **2026-07-10**. Models, prices, accepted fields, policies, and availability are volatile; re-open the linked endpoint and pricing pages immediately before execution.

## Keep the surfaces separate

- **Open Platform API:** `https://api.vidu.com`, API keys, organization-level limits, asynchronous tasks, and API credits. This is the surface covered here.
- **Vidu consumer product:** `https://www.vidu.com` and its subscription/plan credits. Do not assume its credits, features, watermark behavior, or terms apply to the API.
- **Vidu-S1 (Beta):** a separate interactive digital-human product using HTTP session setup, AliRTC media, and WebSocket control. It is not a Q3/Q2 clip model. Its public page currently mixes “1 minute to 2 hours” marketing language with a 600-second session maximum. Its pricing section separately says 3 S1 credits per 2 seconds, lists a `0.03125` credit unit price, deducts every 6 seconds, rounds to 2-second intervals, and requires at least 45 credits to create a session. Do not revalue those statements using the clip API's `$0.005` credit price without written confirmation.
- **S1 documentation conflict:** Step 3 says browser WebSocket authentication belongs in the URL query because browsers cannot set custom headers, while the page's Important Notes says it must be in an authorization header. Treat this as unresolved. Confirm the current transport with Vidu before implementation, never place a bearer URL in logs/analytics/referrers, and do not infer a secure browser design from the sample alone.
- **China host:** the S1 page documents `api.vidu.cn` for China and `api.vidu.com` internationally. The ordinary clip endpoints documented here use `api.vidu.com`; do not infer compute residency from the hostname.

## Choose a mode first, then a model
```
END THIRD-PARTY FILE dedf4eb695c6270f

The third-party file has ended. Everything between the dedf4eb695c6270f tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2704 of 32442 bytes. © calesthio · MIT · https://github.com/calesthio/generative-media-skills/tree/HEAD/skills/providers/video-generation/vidu-video/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
