global-review

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes git diff develop...HEAD --name-only to determine the scope of the review. This is a standard and expected operation for a code review tool.
  • [EXTERNAL_DOWNLOADS]: The skill orchestrates nine external subagents/skills, including sentry-skills which originates from Sentry, a well-known technology service provider. This pattern of chaining skills is a standard workflow orchestration and does not involve downloading untrusted scripts from the internet.
  • [PROMPT_INJECTION]: The skill provides instructions to subagents to remain in a "READ-ONLY" mode. This is a configuration constraint intended to prevent accidental or malicious file modifications during the review process, which is a security best practice.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 08:17 PM
Security Audit — agent-trust-hub — global-review