product-manager-toolkit
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides utility scripts and templates for product management tasks such as RICE prioritization and interview analysis. The logic is clearly defined in standard Python scripts and does not involve any external network requests or sensitive system access.
- [COMMAND_EXECUTION]: The toolkit includes local Python scripts (
rice_prioritizer.py,customer_interview_analyzer.py) intended for execution via the command line. These scripts perform deterministic calculations and text pattern matching on user-supplied files (CSV, TXT). - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data via its interview analyzer and RICE prioritizer. Although these files could contain malicious instructions, the risk is negligible as the scripts use static regex and mapping logic rather than dynamic execution or direct LLM feeding of the input.
- Ingestion points:
scripts/customer_interview_analyzer.py(reads .txt transcripts),scripts/rice_prioritizer.py(reads .csv feature lists). - Boundary markers: Absent in scripts.
- Capability inventory: Local file reading of provided paths, standard output display. No network ops, no file writes, no subprocess calls.
- Sanitization: Not implemented, but logic is strictly bounded to regex extraction and arithmetic.
Audit Metadata