frontend

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

• Third-party content exposure detected (high risk: 0.90). The skill's "Review UI" mode explicitly instructs the agent to WebFetch and "fetch fresh guidelines" from the public URL https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md and to read and apply those fetched rules as part of its review, which means untrusted third-party content could directly influence the agent's actions and outputs.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's "Review UI" mode explicitly instructs fetching runtime content from https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md and using those fetched rules and output-format instructions to control the agent's review behavior, which is a direct runtime dependency that controls prompts/instructions.