Canvas Claw

提供 Python 命令行脚本，向 AI-video-agent 提交图片与视频生成任务，支持文生图、参考图生成、文生视频、首帧生视频，并附带图片/视频模型列举与登录辅助脚本；运行需配置服务地址、站点 ID 与访问令牌。

热度 40.28

最近更新 2026-05-11 17:40:56

分类：创作依赖：代码执行 / 第三方 API / 读写文件应用场景：图像生成 / 视频编辑 / API 集成

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add canvas-claw

Metadata

name: canvas-claw
description: Generate images and videos through AI-video-agent. Supports image create, image remix, video create, and video animate.
metadata: {"openclaw":{"emoji":"🎬","requires":{"bins":["python3"],"env":["AI_VIDEO_AGENT_BASE_URL","AI_VIDEO_AGENT_TOKEN","AI_VIDEO_AGENT_SITE_ID"]},"primaryEnv":"AI_VIDEO_AGENT_TOKEN"}}

Canvas Claw is an OpenClaw skill package that sends generation tasks to AI-video-agent.

图片生成：python3 {baseDir}/scripts/generate_image.py --prompt "cinematic portrait"
参考图生成：python3 {baseDir}/scripts/generate_image.py --prompt "same character, winter coat" --reference-image "/tmp/char.png"
文本生成视频：python3 {baseDir}/scripts/generate_video.py --prompt "a rainy city street at night"
图片生成视频：python3 {baseDir}/scripts/generate_video.py --prompt "the character turns and smiles" --first-frame "/tmp/frame.png"

python3 {baseDir}/scripts/login.py --base-url "http://localhost:8000" --site-id 10000 --username "demo" --password "secret"