mgsci-data-analysis

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's instructions are focused on academic integrity and methodological rigor, such as verifying proofs, identifying causal effects, and ensuring reproducibility.
  • [DATA_EXPOSURE]: The skill addresses the Management Science Code and Data Disclosure Policy, which is a standard academic requirement. It correctly identifies the need for disclosure plans when handling sensitive or proprietary data.
  • [COMMAND_EXECUTION]: The skill references the use of specialized tools (StatsPAI and Stata MCP) for statistical analysis. These tools are used in a controlled manner to automate auditing and sensitivity checks, such as honest_did_from_result and wild_cluster_bootstrap.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 11:57 PM
Security Audit — agent-trust-hub — mgsci-data-analysis