skills/google/adk-python/adk-review/Gen Agent Trust Hub

adk-review

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • Repository Analysis: The skill performs analysis on local code by executing git status and git diff. These are standard operations for identifying changes in a development environment and do not involve network transmission of the code.
  • Development Tool Integration: The workflow includes calls to pytest and pre-commit hooks. These are well-established utilities for automated testing and code linting, and their use here aligns with best practices for maintaining code quality.
  • Review Process and Safety: The skill is designed to generate a report for human review and explicitly requires a user's command before attempting to fix any identified issues. This 'human-in-the-loop' design provides a significant safety barrier against unintended modifications.
  • Contextual Data Handling: The skill processes local file content to perform its review. While processing repository data can involve indirect prompt injection surfaces, the skill's focus on structured reporting and the requirement for explicit authorization for actions mitigate these considerations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 08:11 PM
Security Audit — agent-trust-hub — adk-review