reviewgate-loop

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses git and the GitHub CLI (gh) to perform routine repository tasks such as viewing pull request status, committing fixes, and pushing changes. These operations are standard for the skill's intended purpose and are implemented using authenticated tools.
  • [PROMPT_INJECTION]: The skill processes untrusted data from GitHub pull request comments and tool-generated JSON artifacts. To mitigate indirect prompt injection, it explicitly directs the agent to treat these inputs as evidence for review rather than as instructions to be executed. This aligns with safety best practices for agents ingesting external data.
  • [EXTERNAL_DOWNLOADS]: The workflow mentions the reviewgate-cli and provides instructions for execution using cargo run. This tool is central to the skill's functionality and is referenced in a manner consistent with typical developer tooling usage for the specified vendor.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 01:14 PM
Security Audit — agent-trust-hub — reviewgate-loop