Sora Video Generation

Generate videos using OpenAI's Sora API. Use when the user asks to generate, create, or make videos from text prompts or reference images. Supports image-to-video generation with automatic resizing.

Key points to consider before installing: - This tool will send your prompts and any reference images to OpenAI's video endpoints and requires your OpenAI API key (OPENAI_API_KEY or --api-key). The registry metadata failing to declare this is an inconsistency you should confirm with the publisher before proceeding. - The included script depends on Python packages (openai, httpx, pillow) but there is no install step — install those in a controlled environment (virtualenv) before running. - The script only talks to api.openai.com and saves the returned MP4; it does not contact other endpoints or exfiltrate unspecified data, based on the provided files. - Verify you are comfortable providing your OpenAI API key to this code; consider creating a dedicated API key with limited billing/quota for use with third-party skills. - If you need higher assurance, request the publisher to update registry metadata to declare OPENAI_API_KEY and to provide a reproducible install spec (requirements.txt or install script) and/or review their ownership/source (homepage is missing).

Type: OpenClaw Skill Name: sora-video-gen Version: 1.0.0 The skill bundle is benign. The `SKILL.md` provides clear instructions for generating videos using the OpenAI Sora API without any prompt injection attempts or malicious directives. The `scripts/generate_video.py` script uses standard libraries (`openai`, `httpx`, `pillow`) to interact with the legitimate `api.openai.com` endpoint, handles API keys securely, and performs necessary file operations (reading user-provided images, writing output video, using temporary files) without any evidence of data exfiltration, unauthorized execution, or persistence mechanisms.