Volcengine Doubao Image Gen

豆包图像与视频生成。使用火山引擎 API 生成图片或短视频，支持 Seedream 文生图/图生图/组图，以及 Seedance 文生视频。触发词：豆包图片、doubao、生成图片、火山引擎、字节跳动图片生成、豆包视频、Seedance、生成视频。

This skill otherwise appears coherent for Volcengine image/video generation, but there are two things you should check before installing or running it: 1) The video script will try to load /root/.openclaw/workspace/.env at startup. That file may contain other secrets or tokens from your agent workspace. Inspect that .env (or remove/modify the load_env_file call) so the skill only sees ARK_API_KEY and nothing else. If you don't want the skill to read workspace config, remove or change the hard-coded path in scripts/generate_video.py. 2) The code imports 'volcenginesdkarkruntime' but the package.json and SKILL.md do not declare how to install it. Running the skill as-is may fail or require you to install an SDK from an external source. Verify the correct SDK package name and source (prefer official PyPI package or vendor docs) before installing. Additional cautions: run the skill in an isolated environment (or container) the first time, limit the ARK_API_KEY scope where possible, and confirm network egress is restricted to the documented host (ark.cn-beijing.volces.com). If you need help, ask the skill author to remove the workspace .env read and to add explicit dependency/install instructions.

Type: OpenClaw Skill Name: volcengine-doubao-image-gen Version: 1.1.0 The skill bundle provides legitimate functionality for generating images and videos using the Volcengine (Doubao/Seedance) API. The Python scripts (generate_image.py and generate_video.py) include proactive security measures, such as path traversal checks (ensure_safe_output_path), filename sanitization, and validation of download URLs and content types. The requested network permissions in package.json are limited to the official API endpoint (ark.cn-beijing.volces.com), and no evidence of data exfiltration or malicious prompt injection was found.