receiving-orders

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process external feedback from PR comments and 'external sources', which constitutes a surface for indirect prompt injection. An attacker could embed malicious instructions within a code review comment to influence the agent's behavior. While the skill mandates a 'Verify before implementing' protocol, it lacks explicit boundary markers or automated sanitization for this untrusted input.
  • Ingestion points: PR comments, external reviewer feedback (SKILL.md).
  • Boundary markers: None explicitly defined for untrusted text.
  • Capability inventory: The skill implies file modification capabilities ('IMPLEMENT') and utilizes the GitHub API for thread replies.
  • Sanitization: No validation or escaping of external content is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 07:20 PM