monthly-close
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill handles an ingestion surface for financial data. Ingestion points: Bank statements, email receipts, and physical records are processed by the agent. Boundary markers: The process uses a separate reference file (CLOSE-ARTIFACTS.md) to define closure conditions, acting as a verification boundary. Capability inventory: The skill uses standard file operations and a local Python script for summarization. Sanitization: No explicit data sanitization is defined for incoming financial data strings.
- [COMMAND_EXECUTION]: The script scripts/monthly_close_summary.py is used to aggregate task statuses from JSON files. It takes a file path as an argument and prints counts to the standard output.
Audit Metadata