nmf-topic-modeling
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill provides clear, instructional content for a common data science task.
- [COMMAND_EXECUTION]: The skill includes Python scripts intended for local execution. The code uses well-known libraries such as scikit-learn, numpy, and pandas for text preprocessing and matrix factorization. All operations are local to the document corpus and do not involve suspicious system calls or network activity.
- [DATA_EXFILTRATION]: There are no network requests, credential harvesting, or indicators of data being sent to external servers.
- [PROMPT_INJECTION]: The instructions do not attempt to override agent behavior or bypass safety filters.
Audit Metadata