Story Video Skill

当用户提到以下内容时触发此技能： 把故事做成视频、生成视频、故事视频、AI视频制作、文生视频、图生视频、剧本转视频、分镜生成、分镜头、导演分镜、编剧分析、故事分镜、需要分镜、给视频分镜、拍成视频、拍成分镜、generate video、story to video、video generation、video s...

This skill appears to implement a legitimate story→video pipeline that uses a MiniMax API key. Before installing: 1) Confirm the skill source/trustworthiness (owner slug looks unknown). 2) Inspect SKILL.md and scripts locally for invisible characters (use cat -v, od -c, or a hex editor) and review the Python scripts (they call only MiniMax endpoints and ffmpeg). 3) Resolve metadata mismatches — the registry claims no required env vars but SKILL.md/README require MINIMAX_API_KEY. 4) Run the code in an isolated environment (container or VM) and do not supply high-privilege credentials — provide only a dedicated MiniMax API key with limited quota. 5) Be aware of cost/rate limits described in the README and ensure network access and ffmpeg are configured. If you lack the ability to review code, treat this package as higher risk and avoid granting it credentials or running it on sensitive hosts.

Type: OpenClaw Skill Name: story-video-skill Version: 1.5.1 The skill bundle provides a functional pipeline for AI video generation using the MiniMax API, but it includes high-risk system-level operations. Specifically, 'scripts/merge_videos.py' contains an 'install_ffmpeg' function that attempts to execute 'sudo apt-get install' or 'brew install' to modify the host system's software. Additionally, several scripts ('full_pipeline.py', 'merge_videos.py', etc.) make extensive use of 'subprocess.run' to execute system commands. While these capabilities are plausibly needed for the stated purpose of processing video and managing dependencies, the use of 'sudo' and the potential for system modification through automated scripts represent broad permissions that warrant a suspicious classification.