amnog-millionen-therapie
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely instructional and does not include any executable code, shell scripts, or external dependencies.
- [SAFE]: All external URL references target official European Commission domains (ec.europa.eu) for Health Technology Assessment documentation. These are well-known, trusted sources.
- [SAFE]: The skill encourages the use of established legal databases such as gesetze-im-internet.de and dejure.org for verification, which are standard practices for German legal research.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data (case documents, medical records, legal briefs). While this creates a theoretical attack surface for indirect prompt injection, the skill does not have any dangerous capabilities (like network exfiltration or code execution) enabled, and it includes explicit instructions for data-saving processing of sensitive health information.
- [DATA_EXPOSURE_&_EXFILTRATION]: No evidence of credential harvesting or data exfiltration. The instructions focus on local document analysis and retrieval of official norms.
Audit Metadata