ra-qm-skills
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes 17 Python scripts (e.g.,
risk_matrix_calculator.py,gdpr_compliance_checker.py) to perform regulatory calculations and compliance verification. The author indicates these scripts rely solely on the Python standard library. - [EXTERNAL_DOWNLOADS]: Installation instructions include a command to fetch the skill package from an external GitHub repository using the
agent-skills-clitool. - [PROMPT_INJECTION]: The skill is designed to process and analyze regulatory documents, which inherently creates an indirect prompt injection surface.
- Ingestion points: Processes external regulatory files, QMS documentation, and compliance records.
- Boundary markers: None are defined in the SKILL.md to isolate user-provided data from system instructions.
- Capability inventory: Includes the ability to execute command-line Python scripts for analysis.
- Sanitization: No specific input sanitization or validation mechanisms are described in the provided manifest.
Audit Metadata