code-reviewer

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md explicitly instructs the agent to "Use web_fetch to find Google's specific guidance" and lists public URLs (e.g., google.github.io, skills.sh), so the agent will fetch and interpret open/public third‑party web content during its review workflow which can influence decisions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly directs using web_fetch at runtime to retrieve external guidance (e.g., https://google.github.io/eng-practices/review/reviewer/ and https://skills.sh/google-gemini/gemini-cli/code-reviewer), meaning fetched remote content would directly influence the agent's instructions and behavior.