cs-draft-response

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (customer inquiries) to generate responses, which introduces a potential surface for indirect prompt injection attacks.
  • Ingestion points: Customer inquiry text provided directly or through the cs-ticket-triage workflow in SKILL.md.
  • Boundary markers: The workflow includes explicit instructions for quality checks, tone consistency, and manual review, which act as organizational boundaries.
  • Capability inventory: The skill is restricted to text generation; it lacks network operations, file system write access, or command execution capabilities.
  • Sanitization: The skill mandates the masking of personally identifiable information (PII) such as resident registration numbers, account numbers, and contact details using established rules referenced in references/kr-pii-masking.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 01:41 PM
Security Audit — agent-trust-hub — cs-draft-response