raf-score-calculator
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional Markdown text and contains no executable code, binary assets, or scripts. All described functionality is performed via natural language instructions to the AI agent.
- [EXTERNAL_DOWNLOADS]: The documentation includes a installation command using
npxthat references the vendor's official repository (Open-Medica/open-medical-skills). This is a standard installation procedure for the platform and originates from a recognized vendor source. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process patient encounter documentation and clinical notes from external sources. While this constitutes an attack surface, the skill lacks technical capabilities—such as network exfiltration or file system manipulation—that could be leveraged by a malicious input.
- Ingestion points: Encounter documentation, clinical notes, and patient problem lists (SKILL.md).
- Boundary markers: None explicitly defined.
- Capability inventory: None. The skill is limited to text analysis and calculation within the agent's context.
- Sanitization: None defined.
Audit Metadata