video-setup

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow reads video transcripts from user-provided URLs via yt-dlp to build the voice profile, which is classified as low risk since the agent must actively fetch user-selected URLs rather than monitoring a public feed.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs downloading and executing a remote installation script from an arbitrary third-party domain (https://treg.to/install.sh) during setup, introducing a weak runtime external dependency.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (medium risk: 0.30). The skill sets up video editing tools, folder structures, local configuration files, and references optional external integrations like Treg for auditing/publishing and ElevenLabs/Codex for media generation. While Treg mentions paid API calls ($0.001 a call, balance checking, and free connected accounts), the skill itself does not perform direct payments, transfers, charges, or execute financial instructions. It only configures CLI/MCP access and accounts, leaving API usage management to external services and user actions. Therefore, this is an indirect financial operation/setup context rather than a direct financial execution.