requesting-code-review

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute git commands (git rev-parse, git log, git diff) to identify commit ranges and extract code changes. These are standard and expected operations for a code review workflow.
  • [PROMPT_INJECTION]: The skill interpolates implementation descriptions and code diffs into a template for a subagent, creating a surface for indirect prompt injection. 1. Ingestion points: {WHAT_WAS_IMPLEMENTED}, {PLAN_OR_REQUIREMENTS}, {DESCRIPTION}, and git diff output in SKILL.md and code-reviewer.md. 2. Boundary markers: Absent. 3. Capability inventory: Access to local source code repository via git. 4. Sanitization: Absent. These findings are typical for development-oriented agents and do not indicate malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 02:15 AM