aims-audit
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data via several JSON files through analysis scripts. * Ingestion points: evidence.json, risks.json, audit_scope.json, and program.json. * Boundary markers: Not specified in the script execution instructions. * Capability inventory: The skill executes local Python scripts to analyze data and generate reports. * Sanitization: No explicit sanitization or validation of the input JSON content is documented.
- [COMMAND_EXECUTION]: The skill executes local Python scripts located in vendor and relative directory paths (aims_gap_analyzer.py, ai_risk_register_builder.py, aims_audit_scheduler.py, and cross_framework_mapper.py) to perform compliance auditing tasks.
Audit Metadata