invoice-chase

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The skill consists entirely of Markdown instructions and documentation; it contains no executable code, scripts, or binaries.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests data from external accounting and payment systems.
  • Ingestion points: Financial reports, invoices, and payment history retrieved via platform tools or user-provided files.
  • Boundary markers: No explicit delimiters are specified for the incoming data.
  • Capability inventory: The agent has the capability to send messages via email and messaging tools.
  • Sanitization: This risk is effectively mitigated by mandatory 'approval gates' that require the owner to explicitly review and confirm every draft before any external action is taken.
  • [SAFE]: The skill follows security best practices by enforcing human review for all external communications and verifying recent transactions to ensure data accuracy before contacting clients.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 01:59 AM
Security Audit — agent-trust-hub — invoice-chase