skills/pluginagentmarketplace/custom-plugin-ai-data-scientist/python-programming/Gen Agent Trust Hub
python-programming
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [Prompt Injection] (SAFE): No instructions found that attempt to override agent behavior or bypass safety filters. The skill contains standard instructional language.
- [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file access, or unauthorized network operations detected. Code examples use generic filenames like 'data.csv' and 'database.db'.
- [Obfuscation] (SAFE): Content is clear and readable. No hidden characters, homoglyphs, or encoded strings (Base64/Hex) found.
- [Unverifiable Dependencies] (LOW): Troubleshooting sections mention standard package installation commands for 'pandas' and 'numpy'. These are well-known libraries from trusted registries. No remote script execution patterns (e.g., curl|bash) are present.
- [Indirect Prompt Injection] (INFO): The skill is instructional and shows how to read data from external files (CSV, Excel, JSON). It does not implement automated processing or decision-making based on this data, representing a negligible risk surface.
Audit Metadata