video-creation

用户输入一个选题或口播稿，自动生成完整短视频成片（文案、分镜、数字人口播 + AI 画面混剪）。适用于「一键成片」「根据选题做视频」等场景。当前 skill 已内聚 TTS、数字人视频生成与 AI 文生视频调用能力。

This skill appears internally coherent for generating videos using the chanjing API and local ffmpeg. Before installing: - Understand that scripts, full text/voiceover and audio files will be uploaded to https://open-api.chanjing.cc using the CHANJING_APP_ID/CHANJING_SECRET_KEY you provide — review the provider's privacy, retention, and security practices. - Only supply credentials with the minimum required scope and rotate them if possible; avoid reusing high‑privilege keys. - Run the skill in an isolated environment (container/VM) if you are concerned about exposing sensitive content. The skill will write intermediate media and final mp4 files to local disk — ensure disk space and manage sensitive outputs. - Inspect run_render.py and clients/* locally (they are included) before running; network operations are visible in those files and go to the declared API host (open-api.chanjing.cc). - If you need to prevent autonomous uploads, control when the agent invokes the skill (e.g., require user confirmation) or do not provide credentials until ready to run a job. Overall the package is consistent with its purpose; the primary operational risk is third‑party data transmission and standard local media handling rather than unexplained or unrelated accesses.

Type: OpenClaw Skill Name: video-creation Version: 1.0.2 The video-creation skill bundle is designed to automate the generation of short videos using the Chanjing (chanjing.cc) API. The bundle includes Python clients for authentication, Text-to-Speech, digital human generation, and AI video creation, along with an orchestration script (run_render.py) that utilizes ffmpeg and ffprobe for local media processing. The code uses standard libraries (urllib, subprocess) and follows secure practices, such as passing arguments as lists to subprocess.run to mitigate shell injection risks. The SKILL.md and associated markdown templates provide detailed instructions for the AI agent to ensure content quality and adherence to API constraints without any evidence of malicious prompt injection, data exfiltration, or unauthorized access.