The Agent Skills Directory

[SAFE]: The skill operates entirely within the scope of legal analysis and regulatory compliance. It provides structured guidance on the EU AI Act (VO (EU) 2024/1689) and correctly identifies the implementation timelines, including specific details regarding the exemption of transparency duties from the high-risk delay (Digital Omnibus on AI).\n- [SAFE]: No external code dependencies, package installations, or remote script executions are present in the skill files. All provided links are official EU legal repositories or standard documentation sources.\n- [SAFE]: The skill does not request or process sensitive user data (credentials, keys, or private files). It functions as a text-based decision support tool based on public legal statutes.\n- [SAFE]: The mention of a test script (python3 scripts/eval.py) in the test.md file refers to a standard internal evaluation framework and does not constitute a remote code execution vulnerability for the agent itself.

ki-transparenzpflichten