spec-requesting-code-review
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Git commands, including
git rev-parse,git log, andgit diff, to identify code changes for review. These operations are essential for the primary functionality but involve shell execution with dynamic arguments derived from the repository's history. - [PROMPT_INJECTION]: The sub-agent instructions in
code-reviewer.mdare susceptible to indirect prompt injection as they interpolate untrusted data (implementation summaries and requirements) into the prompt context without sanitization or protective boundary markers. * Ingestion points:code-reviewer.mdvia placeholders{WHAT_WAS_IMPLEMENTED},{PLAN_OR_REQUIREMENTS}, and{DESCRIPTION}. * Boundary markers: Not present. * Capability inventory: Local command execution (git). * Sanitization: None identified in the provided files.
Audit Metadata