code-reviewer
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to process untrusted data from external sources without proper isolation.
- Ingestion points: In the Integration section of
SKILL.md, the skill usesgh pr view <number> --json bodyandgh pr diff <number>to fetch untrusted content. - Boundary markers: The instructions do not include specific delimiters or warnings to ignore instructions that might be embedded within the pull request body or code diffs.
- Capability inventory: Across its scripts and instructions, the skill utilizes command-line tools including
ghandgit. - Sanitization: There is no logic present to sanitize, validate, or escape the content retrieved from external GitHub repositories before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill executes shell commands using
gh(GitHub CLI) andgitto retrieve codebase information and PR data. While these are appropriate for a code review tool, they are executed in a context where they interact with external, untrusted data.
Audit Metadata