market-research-reports
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The bundled Python scripts implement robust security controls, including symlink rejection, file size limitations (5MB), and path resolution checks to prevent traversal. All operations are confined to the local filesystem.
- [COMMAND_EXECUTION]: The skill provides command-line tools for market sizing and report scaffolding. These scripts are written using the Python standard library and explicitly avoid dangerous operations like dynamic code evaluation, serialization with pickle, or invoking external shell commands.
- [EXTERNAL_DOWNLOADS]: The skill's documentation contains links to official government and international organization data repositories (e.g., Census Bureau, World Bank). These references are for informational routing to well-known, trusted services and do not involve automated downloading of executable code.
- [INDIRECT_PROMPT_INJECTION]: The workflow involves ingesting external data ledgers. The skill mitigates potential data poisoning and indirect injection through mandatory claim-to-source mapping, strict schema validation for CSV/JSON inputs, and clear instructional boundaries that separate sourced facts from AI-generated simulations.
Audit Metadata