NYC

creating-financial-models

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • PROMPT_INJECTION (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) because it is designed to ingest and process external financial statements and models provided by users.
  • Ingestion points: Historical financial statements, acquisition models, and user-defined scenario parameters.
  • Boundary markers: Absent. The instructions do not specify delimiters or warnings for the agent to ignore instructions embedded within the processed financial data.
  • Capability inventory: Executes local Python scripts (dcf_model.py, sensitivity_analysis.py) to perform calculations and generate Excel workbooks.
  • Sanitization: Absent. No explicit validation or filtering of external content is mentioned.
  • Risk Assessment: Since the output is primarily for agent reasoning and local file generation (Excel), the severity is classified as LOW.
  • EXTERNAL_DOWNLOADS (INFO): The skill references external Python scripts and is part of a remote repository.
  • Source: The metadata points to the anthropics GitHub organization.
  • Trust Status: Anthropic is a Trusted GitHub Organization. Per [TRUST-SCOPE-RULE], the finding regarding remote source references is downgraded to INFO.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 02:04 AM