skills/lawve-ai/awesome-legal-skills/eu-ai-act-classification-oliver-schmidt-prietz/Gen Agent Trust Hub
eu-ai-act-classification-oliver-schmidt-prietz
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill uses a structured logical workflow to ensure consistent analysis. While it ingests user-provided text (Phase 1 Q1 and Prior Assessment Context in SKILL.md), which constitutes an attack surface for indirect instructions, the agent's capabilities are limited to informational searches. There are no high-privilege tools like file-system access or code execution, rendering the risk negligible.
- [SAFE]: No obfuscation, Base64 encoding, or hidden characters were detected across the skill's instructions or reference files. All URLs are clearly visible and point to relevant project documentation.
- [DATA_EXFILTRATION]: No sensitive data access or unauthorized network requests were found. The skill suggests performing web searches for official regulatory updates, which is standard and safe behavior for its purpose.
- [REMOTE_CODE_EXECUTION]: The skill contains no executable scripts or commands to download and run remote code. It functions purely as a set of reasoning instructions for the AI agent.
Audit Metadata