requesting-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external content by interpolating project plans and implementation summaries into a subagent prompt. * Ingestion points: The [DESCRIPTION] and [PLAN_OR_REQUIREMENTS] placeholders within the code-reviewer.md template. * Boundary markers: The subagent prompt uses Markdown headers to delimit sections of input data. * Capability inventory: The subagent is authorized to use git commands (diff, log, show, worktree) for code analysis. * Sanitization: The skill does not implement specific filtering or escaping for the interpolated content.
- [COMMAND_EXECUTION]: The workflow involves executing standard git commands like rev-parse, log, and diff, which are necessary for identifying commit ranges and comparing code versions.
Audit Metadata