payment-failure-diagnosis

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill consists entirely of markdown instructions and does not include any scripts, executable files, or tool definitions. All logic is prompt-based and intended for human or AI reasoning within the chat interface.
  • [SAFE]: No patterns of prompt injection, data exfiltration, or obfuscation were detected. The skill uses standard financial industry terminology and codes (NACHA, SWIFT, Fedwire) to guide the analysis of payment exceptions.
  • [SAFE]: The skill does not perform any network operations, access sensitive local system files, or include hardcoded credentials. It defines a data ingestion schema for manual input but does not automate data collection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 03:14 AM
Security Audit — agent-trust-hub — payment-failure-diagnosis