ai-act-use-case-screening
Preliminary EU AI Act screening of any AI use case, planned or already in use, from the user's own description, documents and answers. Asks questions in short batches, then writes a structured report with the risk classification in one of the four AI Act tiers (unacceptable risk, high-risk, limited risk with transparency obligations, minimal or no risk), what the user must do and by when, optional recommendations, the dates that affect them, what would change the classification, and open questions. For a system already in use, every required action gets a status (in place, partly, missing, unknown). Every finding carries evidence from the user's words or documents, an AI Act citation and the reasoning. Use when someone wants to know whether and how the EU AI Act applies to an AI system or idea, asks for its risk class, or wants an AI Act gap check. For a GAIK Solution Configuration Wizard pack, use ai-act-screening instead.
- Compatibility
- Needs Python 3.10 or later (standard library only) to run scripts/check_screening.py, and file access for the screening folder. Runs in claude.ai with code execution enabled, Claude Code, and other agents that can run Python.
Pinned to revision 212201f203cb, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/ai-act-use-case-screening/SKILL.md
- skills/ai-act-use-case-screening/README.md
- skills/ai-act-use-case-screening/references/ai-act-reference.md
- skills/ai-act-use-case-screening/references/output-format.md
- skills/ai-act-use-case-screening/references/question-bank.md
- skills/ai-act-use-case-screening/references/report-guide.md
- skills/ai-act-use-case-screening/scripts/check_screening.py
Every link opens the file at its source, pinned to the revision this page describes.