Jimeng Api Video Gen

This connects Claude to Volcengine's Jimeng Video 3.0 API for AI video generation from text prompts or images. You get multiple quality tiers (720p, 1080p, and Pro), adjustable duration from 4 to 10 seconds, and different modes including text-to-video, image-to-video with first frame control, and camera movement options. The implementation handles Volcengine's IAM v4 signature auth and async polling automatically. Generation takes 2-10 minutes and returns a 24-hour download URL. Useful if you're building video creation tools or need programmatic access to Jimeng's models without handling the API complexity yourself. You'll need Volcengine credentials and should watch out for content moderation failures and rate limits.

skills.sh installs: 481
GitHub stars: 1
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add wvlvik/toy-skills --skill jimeng-api-video-gen --agent claude-code

Installs into .claude/skills of the current project.

Files

Jimeng API Video Generation

Generate high-quality AI videos using Volcengine's Jimeng Video 3.0 Pro API with text prompts or input images.

Configure Environment Variables

Get Access Key/Secret Key

# ~/.zshrc or ~/.zshenv
export VOLC_ACCESSKEY="your_access_key_id"
export VOLC_SECRETKEY="your_secret_access_key"

Output Behavior (Important)

After video generation succeeds, you MUST display the video directly to the user!

When generation completes, output in the following format:

```markdown Generated Video:

Download Video

Generation Info:

Prompt: [User's original prompt]
Duration: [Video length in seconds]
Mode: [Generation mode used]
FPS: [Frames per second]
Save Path: [Local file path, if applicable] ```

Example Output

```markdown Generated Video:

Download Video

Generation Info:

Prompt: A golden retriever running through a field of sunflowers, warm afternoon sunlight
Duration: 5 seconds
Mode: t2v_720p
FPS: 24
Save Path: /generated_videos/dog_sunflowers.mp4