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 uses local git commands to manage the code review process.
  • Evidence: Executes git rev-parse, git log, and git diff within SKILL.md and code-reviewer.md to identify changes and generate diff statistics.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it processes external content.
  • Ingestion points: Untrusted data from commit descriptions ({DESCRIPTION}), requirements ({PLAN_REFERENCE}), and the output of git diff are interpolated directly into the subagent prompt in code-reviewer.md.
  • Boundary markers: None. There are no delimiters (e.g., XML tags or triple quotes) or 'ignore' instructions surrounding the variables to prevent the LLM from following instructions embedded within the code or descriptions.
  • Capability inventory: The skill has the ability to execute shell commands via git.
  • Sanitization: None. The skill does not sanitize or escape the content of the diff or the implementation details before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 04:19 AM