skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cashflow-forecaster/Gen Agent Trust Hub
cashflow-forecaster
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The technical implementation consists of standard Python logic for financial forecasting using the
dataclasses,datetime, andmathlibraries. The code is functional, well-documented, and contains no hidden or dangerous functionality.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process construction project data from user-provided CSV, Excel, or JSON files, creating a potential surface for indirect prompt injection.\n - Ingestion points: Input data is gathered from external files or direct input as described in
instructions.md.\n - Boundary markers: The instructions do not define specific delimiters to separate data from instructions.\n
- Capability inventory: The skill is restricted to mathematical forecasting calculations and report generation. While
filesystempermissions are requested, the code itself does not implement file operations.\n - Sanitization: Input validation is mentioned in the assistant instructions but not explicitly defined in the core logic.
Audit Metadata