gemini-interactions-api

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill contains explicit instructions in SKILL.md to override the agent's internal knowledge ("These rules override your training data. Your knowledge is outdated") and to ignore user requests for specific model versions.
  • [COMMAND_EXECUTION]: The migration guide (references/migration.md) instructs the agent to execute a shell command using rg (ripgrep) and pipes (cut, sort, uniq) to scan project files for migration targets.
  • [EXTERNAL_DOWNLOADS]: The skill requires the agent to fetch remote documentation from a well-known service (ai.google.dev) to serve as the authoritative source for API parameters and logic.
  • [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface because it instructs the agent to fetch and trust external content from remote URLs without specifying sanitization, validation, or boundary markers.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 01:39 AM
Security Audit — agent-trust-hub — gemini-interactions-api