skills/mr-kelly/skills/kelly-finance/Gen Agent Trust Hub

kelly-finance

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script (scripts/build_three_statement_model.py) via the Node.js spawn function. This is performed securely using an arguments array rather than a shell string, preventing shell injection vulnerabilities. This operation is limited to generating financial workbooks locally.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes financial data retrieved from Busabase and rendered in a web UI. While this represents a potential ingestion point for indirect instructions, the risk is mitigated by the use of structured data (JSON), numerical validation, and explicit human-in-the-loop review steps defined in the workflow.
  • [DATA_EXPOSURE]: Configuration and secrets are managed via environment variables (e.g., BUSABASE_API_KEY) rather than being hardcoded in scripts or the application. The skill is scoped to interact only with the user's provided Busabase workspace.
  • [DYNAMIC_EXECUTION]: The web UI uses dynamic imports to switch between live data and demo modes. This is a standard architectural pattern for modern web applications and does not involve the execution of untrusted or remote code.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 01:36 PM