Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Sora Video
v1.0.0Generate, edit, extend, and manage AI videos using OpenAI's Sora 2 API. Includes marketing-ready prompt templates for product demos, social ads, brand spots,...
⭐ 0· 61·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description match the bundled CLI and templates: the skill legitimately needs an OpenAI API key and the openai SDK to call Sora. However, the registry metadata lists no required environment variables while SKILL.md and the CLI clearly require OPENAI_API_KEY; that metadata omission is an inconsistency.
Instruction Scope
Runtime instructions instruct the user to export OPENAI_API_KEY, run the bundled Python CLI to create/edit/extend videos, and upload local media as input references (which will be sent to OpenAI). This is coherent with purpose, but the instructions also suggest installing 'uv' via a remote shell installer (curl -LsSf https://astral.sh/uv/install.sh | sh). Uploads of local files to the API and running a remote install script are the primary operational risks and should be explicitly acknowledged by the user.
Install Mechanism
There is no formal install spec in the registry, but SKILL.md recommends installing 'uv' by piping a remote install script from astral.sh into sh — a high-risk pattern because it executes third-party code fetched at runtime. The skill relies on uv to auto-install the openai SDK; that remote-install recommendation increases attack surface and warrants manual review or safer installation alternatives.
Credentials
The only sensitive credential the skill needs is OPENAI_API_KEY (required by the SKILL.md and used by the CLI). That is proportionate to the stated purpose. Still, the registry metadata not declaring this required env var is inconsistent and should be corrected. The SKILL.md explicitly warns not to paste keys into chat, which is good.
Persistence & Privilege
Skill flags show no always:true and normal autonomous invocation defaults. The skill does not request persistent system-wide privileges in the manifest. The usual agent-autonomy caveats apply but there are no elevated privileges requested by the skill itself.
What to consider before installing
What to consider before installing:
- The skill is coherent with its purpose (Sora video generation) and legitimately requires an OpenAI API key; expect your videos and any reference files you upload to be sent to OpenAI's API. Only supply media you are comfortable sharing.
- Do NOT run arbitrary curl | sh installers without review. SKILL.md suggests running a remote installer (astral.sh) to get 'uv'. Prefer installing uv via a trusted package manager or inspect the install script before executing it. Running an unreviewed remote install elevates risk.
- Confirm the registry metadata is updated to list OPENAI_API_KEY as a required env var. The current mismatch is a red flag about maintenance quality.
- Limit the API key's exposure: use a key tied to a billing account you control; enable usage limits if possible and monitor billing for unexpected usage, since video generation can be costly.
- If you need stronger assurance, ask the publisher for: source code repo/homepage, checksum or signed release for the CLI, and an explanation why there is no install spec in the registry. Running the CLI in an isolated environment (non-root container or ephemeral VM) until you vet it is recommended.
- If you plan to upload sensitive assets, verify that the guardrails (no real people, no copyrighted music/characters) match your requirements and consider pre-processing or anonymizing inputs.
What would change this assessment: presence of an official homepage or code repository, a registry install spec that uses vetted package sources (instead of recommending curl|sh), and corrected metadata explicitly listing OPENAI_API_KEY would move this toward 'benign'. Conversely, evidence of hidden endpoints, telemetry, or requests for unrelated credentials would increase suspicion.Like a lobster shell, security has layers — review code before you run it.
latestvk970h64hqq6s3rhsx97xghcwpn838yq2
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
