people-analytics
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes workforce metrics and engagement survey results from external CSV files. This creates a surface where malicious instructions could be embedded in data fields (such as job descriptions, survey comments, or department names) to influence the agent's analysis or recommendations.
- Ingestion points:
scripts/attrition_predictor.py,scripts/headcount_planner.py, andscripts/survey_analyzer.pyread untrusted data from CSV files specified via the--fileargument. - Boundary markers: Absent. The scripts do not implement specific delimiters or instructions to help the agent distinguish between legitimate data and potential instructions embedded within the processed CSV content.
- Capability inventory: All scripts are restricted to local file reading and statistical calculations. No network operations, subprocess execution, or dynamic code evaluation (eval/exec) were detected in the provided code.
- Sanitization: The scripts use standard CSV parsing and basic type conversion (float/int) for metric calculations, but do not provide sanitization against prompt injection patterns in textual data fields.
Audit Metadata