videogencn — Claude Code video generation skill: install guide
Generate clips from text, images, frame pairs, or references with Chinese video models. Use the skill to animate a still or shape a transition between two frames.
npx skills add Agents365-ai/365-skills --skill videogencn -a claude-codeWhat videogencn does
videogencn selects a generation mode from the materials you provide: a text prompt, an image, a start and end frame, or reference images. It connects to models through Bailian, Jimeng, MiniMax, and Hunyuan. For text-to-video and image-to-video, the skill can refine a prompt and offer alternatives before generation; frame-pair and reference modes skip that step.
The script submits each job, checks its progress, then saves the MP4. Result links expire after 24 hours, so the video is downloaded as soon as the job finishes. You need python3 and an API key for the provider you choose; Bailian uses DASHSCOPE_API_KEY. Some Bailian models are limited to the cn region, which affects model choice.
What it helps with
- generate a clip from a written prompt
- animate a still image with a video model
- create a transition between two supplied frames
- use reference images to maintain subject consistency
Example prompts for the agent
- Generate a short clip from this idea: a night train crossing a snowy forest.
- Animate
cover.pngwith a slow camera move toward the lighthouse. - Create a cinematic transition from
start.pngtoend.png.
How to install videogencn
1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill videogencn -a claude-code2Manually: put the files in place
~/.claude/skills/videogencn/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/videogencn/
1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill videogencn -a codex2Manually: put the files in place
.agents/skills/videogencn/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill videogencn -a cursor2Manually: put the files in place
.cursor/skills/videogencn/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Agents365-ai/365-skills videogencn2Manually: put the files in place
.github/skills/videogencn/SKILL.md1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill videogencn -a gemini-cli2Platform CLI
gemini skills install https://github.com/Agents365-ai/365-skills --path plugins/videogencn/skills/videogencnManually: put the files in place
.devin/skills/videogencn/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill videogencn -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about videogencn
How do I install videogencn in Claude Code?
Run npx skills add Agents365-ai/365-skills --skill videogencn -a claude-code. The skill file is placed at ~/.claude/skills/videogencn/SKILL.md.
What does videogencn do?
It creates video from text, an image, a pair of frames, or reference images using supported Chinese video models. The script retrieves the finished result as an MP4.
What does videogencn need to run?
You need python3 and an API key for the selected provider. Bailian requires DASHSCOPE_API_KEY, and its PixVerse, Kling, Vidu, and HappyHorse models are restricted to the cn region.
