skills/lawve-ai/awesome-legal-skills/eu-ai-act-obligations-oliver-schmidt-prietz/Gen Agent Trust Hub
eu-ai-act-obligations-oliver-schmidt-prietz
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely informational and analytical in nature. It provides guidance on the EU AI Act (Regulation (EU) 2024/1689) by processing user input and referencing internal documentation.
- [PROMPT_INJECTION]: No malicious prompt injection patterns were found. The instructions focus on structured workflows, adaptive intake, and generating compliance matrices. Standard behavioral instructions like 'Ask Questions ONE AT A TIME' are benign operational constraints.
- [DATA_EXFILTRATION]: No sensitive file access, credential exposure, or unauthorized network operations were detected. The skill uses web search capabilities to find the latest regulatory updates, which is consistent with its stated purpose as a compliance tool.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user descriptions of their AI systems. While this presents a theoretical attack surface, the skill's capabilities are limited to text generation and web searching. It does not execute code or perform high-risk actions based on the ingested data, making the risk negligible.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any remote code execution (RCE) or download executable scripts. It includes a link to a GitHub Pages site for an interactive version of the tool, which is a legitimate and transparent use of external resources.
Audit Metadata