requesting-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including
git rev-parse,git log, andgit diffto identify commit SHAs and retrieve code changes for review. These are standard operations for the skill's intended purpose. - [INDIRECT_PROMPT_INJECTION]: The skill processes implementation descriptions and plan requirements which are interpolated into the subagent's instructions, creating a surface for indirect prompt injection.
- Ingestion points: The
code-reviewer.mdtemplate populates placeholders such as{WHAT_WAS_IMPLEMENTED},{PLAN_OR_REQUIREMENTS}, and{DESCRIPTION}with data from the current task. - Boundary markers: The template does not utilize specific delimiters or instructions to ignore embedded commands within the interpolated data.
- Capability inventory: The skill possesses the capability to execute shell commands (
git) and perform file system analysis. - Sanitization: There is no evidence of sanitization or validation for the content being interpolated into the review template.
Audit Metadata