segmentation-analysis
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a structured workflow for data analysis using standard libraries. All operations, including file reading, clustering, and plotting, are performed locally on the user's data. No network requests or unauthorized system modifications were detected.- [PROMPT_INJECTION]: The skill ingests data from an external file ('user_data.csv'), which is a potential surface for indirect prompt injection if the resulting analysis or data previews are interpreted as instructions by the agent. * Ingestion points: 'user_data.csv' read via pandas in SKILL.md. * Boundary markers: None present in the prompt templates. * Capability inventory: Local file system read/write, data visualization. * Sanitization: Includes data cleaning for analysis (NA removal, outlier detection) but lacks specific sanitization for embedded natural language instructions.
Audit Metadata