financial-modeling
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No instructions to override agent behavior, bypass safety protocols, or extract system prompts were detected.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No hardcoded credentials, sensitive file access, or unauthorized network activity was found; the skill uses office-mcp tools for intended financial modeling purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill processes historical financial data provided by the user (Ingestion points: Excel data via read_xlsx). Capabilities include generating spreadsheet projections and charts (Capability inventory: create_xlsx, apply_formula, create_chart). While no explicit boundary markers or sanitization are defined, the activity is restricted to spreadsheet modeling.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: No external code dependencies, package installations, or remote script execution patterns are present.
Audit Metadata