bayesian-workflow
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Analysis of the instructions and logic in SKILL.md and README.md confirms that the skill adheres to legitimate educational and functional guidelines for Bayesian modeling without attempting to override agent safety protocols.
- [SAFE]: The provided utility scripts, 'diagnose_model.py' and 'calibration_check.py', utilize standard scientific libraries for data processing and visualization.
- Data ingestion is performed using established NetCDF loaders from the ArviZ library, which is a safe practice for handling structured model output.
- File system operations are transparent and limited to reading user-provided input data and writing requested diagnostic reports or plot images.
- [SAFE]: No obfuscation, data exfiltration, or unauthorized remote code execution patterns were found. The skill does not perform suspicious network requests or access sensitive system paths.
- [SAFE]: All external references point to reputable open-source repositories and documentation sites (e.g., PyMC, ArviZ, and GitHub organizations associated with Bayesian statistics).
Audit Metadata