runway-calculator
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a standard financial utility and shows no signs of malicious intent or dangerous behavior.\n- [COMMAND_EXECUTION]: The skill executes a localized Python script (
scripts/runway.py) to perform calculations. The script is restricted to the Python standard library and does not perform any network operations, file writing, or persistent system modifications.\n- [PROMPT_INJECTION]: The skill processes user-provided financial data, which is a potential surface for indirect prompt injection.\n - Ingestion points: The
runway.pyscript reads input data from JSON files or standard input as specified by the user.\n - Boundary markers: Instructions require the agent to format output into specific Markdown tables, helping to maintain a boundary between tool output and agent reasoning.\n
- Capability inventory: The skill's capabilities are limited to executing the provided mathematical computation script.\n
- Sanitization: Input is strictly parsed as JSON and numeric fields are validated through float conversion, which prevents the execution or interpretation of malicious non-numeric instructions.
Audit Metadata