gemini-api-dev

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Contextual Directives: The skill uses explicit instructions to prioritize its provided content over the model's internal training data. This is a common method to ensure accuracy for rapidly evolving technologies, but it represents a control pattern that overrides default agent behavior.
  • External Resource Ingestion: The skill is designed to process content from an external documentation site. This ingestion creates a surface for indirect prompt injection, where instructions embedded in external files could influence agent actions, although the source is the official documentation domain.
  • Package Registry Dependencies: The instructions recommend installing software packages from public registries. These libraries are part of the vendor's official SDK offerings and represent standard development dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 09:03 PM