revenue-recognition

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns such as prompt injection, obfuscation, or data exfiltration were detected. The skill instructions follow legitimate accounting workflow logic.
  • [DATA_EXPOSURE]: The skill operates on local files within the project environment (e.g., workpapers/transactions.json, payee_map.json) and does not perform network operations or access sensitive system paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from workpapers/transactions.json. While this constitutes a potential injection surface for malicious transaction descriptions, the skill does not grant the processed data access to high-privilege capabilities or execute code derived from that data.
  • [COMMAND_EXECUTION]: References to a 'classify script' appear to be internal tool calls within the agent's established environment rather than arbitrary shell execution or remote script downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 02:14 PM
Security Audit — agent-trust-hub — revenue-recognition