survey-analyzer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- SAFE (SAFE): No malicious patterns, prompt injections, or unauthorized network behaviors were detected. The skill primarily uses standard data science libraries for statistical calculations.
- NO_CODE (SAFE): No executable Python code was provided for the implementation of the survey_analyzer.py script, but the described API and CLI behavior align with safe data processing practices.
- Indirect Prompt Injection (SAFE): Ingestion points: load_csv in SKILL.md. Boundary markers: Absent. Capability inventory: File-write (generate_report, plot_likert). Sanitization: Not explicitly shown in documentation. The skill performs programmatic analysis (e.g., sentiment scoring via textblob) rather than instruction-based processing.
Audit Metadata