receiving-code-review
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard command-line tools for development workflows. It uses
grepto verify code usage (YAGNI principle) and the GitHub CLI (gh api) to respond to pull request comments. These operations are within the scope of code review tasks. - [PROMPT_INJECTION]: The instructions contain phrases like "NEVER" and "INSTEAD" to define response patterns (e.g., avoiding "Great point!"). These are legitimate stylistic and behavioral constraints for the agent's persona and do not attempt to bypass platform safety guidelines or extract system prompts.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external input in the form of code review comments. It implements defensive measures by instructing the agent to treat external feedback as suggestions to be evaluated rather than orders, requiring technical verification against the codebase before any implementation. This posture helps mitigate the risk of malicious or incorrect suggestions provided by external reviewers.
Audit Metadata