finance-skills
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill references and provides command-line examples for executing local Python scripts (e.g.,
ratio_calculator.py,dcf_valuation.py,metrics_calculator.py) located within the skill's own subdirectories. These tools are used for specialized financial computations and are documented as having no external dependencies beyond the Python standard library.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external financial data, establishing an attack surface for indirect prompt injection.\n - Ingestion points: Data is read from local JSON files, such as
sample_financial_data.json,valuation_data.json, andbudget_data.json.\n - Boundary markers: The skill includes explicit instructions for the agent to validate the integrity of input data (checking for missing or implausible values) and to verify calculation results against sanity limits and comparable benchmarks.\n
- Capability inventory: The agent is authorized to read local files, execute local scripts, and generate detailed financial reports.\n
- Sanitization: The workflow mandates data integrity validation and manual-style review of outputs as part of the primary analysis phases.
Audit Metadata