vpe-review
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to run local Python scripts (
delivery_throughput_analyzer.py,eng_hiring_funnel_calculator.py, andeng_team_structure_designer.py) located in a relative directory structure (../../../skills/vpe-advisor/scripts/). These scripts are used to process metrics and hiring data. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external files without explicit sanitization or boundary markers.
- Ingestion points:
sprint_metrics.json,funnel.json, andteam.jsonfiles read via the workflow section. - Boundary markers: No delimiters or instructions are used to isolate the data from the prompt context.
- Capability inventory: Execution of local Python scripts with file arguments.
- Sanitization: No evidence of validation or filtering for the input JSON files.
Audit Metadata