gemini-api

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation and model metadata from official Google domains (ai.google.dev and cloud.google.com), which are well-known and trusted services.
  • [COMMAND_EXECUTION]: Instructions include installing official, widely-used SDKs (@google/generative-ai and @google-cloud/vertexai) through standard package managers (pnpm).
  • [CREDENTIALS_UNSAFE]: The skill demonstrates security best practices by instructing the user to store API keys in .env.local files rather than hardcoding them in the source code.
  • [PROMPT_INJECTION]: The skill ingest documentation from trusted sources and user-provided code files to assist with development; while this represents an indirect injection surface, it is necessary for the skill's primary function and lacks malicious intent markers.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 02:13 PM
Security Audit — agent-trust-hub — gemini-api