vercel-sandbox

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Source Initialization" section shows the agent can clone arbitrary git repos and mount tarball URLs (e.g., https://github.com/user/repo and https://example.com/project.tar.gz), meaning it ingests and executes untrusted third-party content that can influence runtime behavior.