public-equity-investing

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • Internal Script Execution: The skill utilizes python3 commands to execute bundled utility scripts for deterministic operations, such as data preparation (clean_tabular_data.py) and report generation (render_dashboard.py). These executions are used for their intended functional purposes and are scoped to the skill's environment.
  • Data Ingestion and Processing: A core function of the skill involves processing user-provided financial datasets in formats like Excel, CSV, and JSON. The skill includes extensive internal policies (e.g., financial-source-of-truth) that guide the agent to strictly separate verified facts from management claims and assumptions, reducing risks associated with processing external data.
  • Metadata and Style Extraction: The skill includes a metadata extraction utility for Office files (extract_office_style.py) which uses standard libraries to read file properties without modifying original documents. This process is used to align deliverables with institutional style guidelines.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 06:13 AM
Security Audit — agent-trust-hub — public-equity-investing