jf-empirical-design
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill is a legitimate academic research tool. No evidence of prompt injection, data exfiltration, obfuscation, or persistence mechanisms was found. The instructions align with professional financial econometrics standards.- [COMMAND_EXECUTION]: The skill incorporates an execution bridge that leverages Model Context Protocol (MCP) tools (specifically
mcp__stata-mcp__stata_do) and Python environments to perform statistical regressions and multiple-testing corrections. These tools are used as intended for research purposes, utilizing internal code skeletons provided in the environment.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided financial data and predictors for statistical validation. - Ingestion points: User-supplied candidate predictors, anomalies, and characteristic signals.
- Boundary markers: The skill does not implement specific boundary markers or "ignore previous instructions" warnings when processing these inputs.
- Capability inventory: The skill possesses the ability to execute code via Stata MCP and Python, accessing local code templates.
- Sanitization: No explicit sanitization or input validation logic is defined for the financial data processed, which is expected for a statistical analysis tool.
Audit Metadata