scanpy
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No instructions were found that attempt to override agent behavior or bypass safety filters.
- [Data Exposure & Exfiltration] (SAFE): All file operations are local. No network access or access to sensitive system paths was identified.
- [Remote Code Execution] (SAFE): No dynamic execution of remote or untrusted code was found.
- [Indirect Prompt Injection] (SAFE): The skill processes biological data files (H5AD, CSV, MTX) as structured numerical data. 1. Ingestion points: assets/analysis_template.py and scripts/qc_analysis.py. 2. Boundary markers: Absent. 3. Capability inventory: Local file I/O and matplotlib plotting. 4. Sanitization: Handled by scanpy and pandas data parsers.
Audit Metadata