receiving-code-review
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill provides instructions to override standard conversational patterns, such as performative politeness (e.g., "You're absolutely right"), in favor of technical rigor. It also defines a specific phrase, "Strange things are afoot at the Circle K," to be used as a signal for technical disagreement. These are categorized as instructional constraints rather than malicious overrides.
- [COMMAND_EXECUTION]: The skill references the use of GitHub CLI commands (
gh api) as the intended method for replying to pull request comments. This usage is transparent and tied to the skill's primary function of managing code reviews. - [SAFE]: The skill's logic is fundamentally defensive. It specifically warns against "blind implementation" and instructs the agent to treat external feedback as suggestions requiring evaluation rather than commands requiring absolute obedience, which mitigates risks associated with indirect prompt injection from malicious code reviewers.
Audit Metadata