gemini-interactions-api

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The list is mostly official ai.google.dev documentation (low risk), but it includes a generic GitHub repository reference (https://github.com/my-org/backend) which is an untrusted/placeholder repo used as an external source and could be used to distribute arbitrary code—this is a suspicious download/source indicator.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires fetching external Gemini documentation at runtime (e.g., https://ai.google.dev/gemini-api/docs/interactions.md.txt) and mandates using that fetched content as the source-of-truth to drive code and agent behavior, so these ai.google.dev URLs are runtime dependencies that directly control the agent's instructions.