chatgpt-video-editing-setup

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

• Third-party content exposure detected (high risk: 0.75). Runbook-driven runtime cloning/registration of third-party repositories (e.g., https://github.com/browser-use/video-use.git and optional https://github.com/heygen-com/hyperframes.git) can bring in outsider-authored text like SKILL.md/other repository prose into the agent’s LLM context via Skills registration, creating an indirect prompt-injection risk.

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

• Potentially malicious external URL detected (high risk: 0.90). The runbook includes runtime git clone and install commands that fetch and install remote code (which will be executed), specifically https://github.com/browser-use/video-use.git and https://github.com/heygen-com/hyperframes.git, which are required dependencies for the Skill.