causal-scientist
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides templates and best practices for causal inference analysis using well-known scientific libraries.
- [EXTERNAL_DOWNLOADS]: The skill uses industry-standard Python libraries such as pandas, numpy, dowhy, networkx, and causal-learn.
- [INDIRECT_PROMPT_INJECTION]: The skill defines ingestion surfaces for external data. 1. Ingestion points: Data is processed through Pandas DataFrames in methods like estimate_effect and discover in references/patterns.md. 2. Boundary markers: No explicit delimiters or instructions are provided to the model to ignore embedded data instructions. 3. Capability inventory: Analysis of all patterns reveals no high-risk capabilities such as file system writes, network requests, or subprocess execution. 4. Sanitization: No explicit sanitization or validation of data content is implemented beyond standard library loading. The lack of dangerous capabilities effectively mitigates the risk of indirect prompt injection.
Audit Metadata