customer-support-verification

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and verify customer-provided data, such as support requests and Gmail thread content, which is a known vector for indirect prompt injection.
  • Ingestion points: Customer requests and Gmail thread/draft bodies are defined as required inputs in SKILL.md.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' prefixes when processing customer-supplied content.
  • Capability inventory: The skill is used to verify actions across high-privilege surfaces including Gmail, Stripe, Firebase, and Git commits.
  • Sanitization: No specific sanitization or validation rules are provided for handling the contents of customer emails.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 01:59 PM
Security Audit — agent-trust-hub — customer-support-verification