code-review-quality

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted source code from external pull requests, creating a surface for indirect prompt injection.\n
  • Ingestion points: Input code is accepted via parameters like prFiles and prNumber as demonstrated in SKILL.md.\n
  • Boundary markers: The skill lacks explicit instructions or delimiters to isolate the reviewed code from agent instructions, which could allow embedded commands in code comments to influence the agent.\n
  • Capability inventory: The skill is capable of spawning sub-tasks and coordinating multiple agents (e.g., security and performance agents) using FleetManager in SKILL.md.\n
  • Sanitization: No evidence of code sanitization or safety-filtering of the input files is provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 07:54 AM