code-review

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external code as input, which presents a surface for indirect prompt injection. A malicious actor could embed instructions within code comments or string literals to bypass security checks or influence the agent's final verdict.
  • Ingestion points: Step 2 of the review process in SKILL.md requires the agent to read each changed file fully.
  • Boundary markers: The instructions lack explicit delimiters or system-level warnings to distinguish the code being analyzed from the agent's core instructions.
  • Capability inventory: The skill's capabilities are restricted to generating text-based review reports. It does not have access to the file system (beyond reading), network operations, or subprocess execution.
  • Sanitization: No sanitization or validation logic is defined to filter or escape potential injection patterns within the ingested code.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 03:53 PM