Static scanner and Annex IV drafter for EU AI Act compliance (prototype)
Project description
🇪🇺 AI Act Compliance Scanner (Sovereign Code)
Don't let a €35M fine stop you from shipping. The open-source CLI that maps your Python code to the EU AI Act (Annex IV) in seconds.
⚡ The Problem
The EU AI Act is here. If your software uses ML libraries (torch, sklearn, face_recognition), you might be classified as "High Risk" under Article 6.
- Lawyers cost: €500/hour.
- Ignorance costs: Up to 7% of global turnover.
- Manual compliance: Boring, error-prone, and slow.
🛡️ The Solution
ai-act-check is a "Compliance-as-Code" tool. It parses your repository's Abstract Syntax Tree (AST), identifies regulated libraries, and uses an LLM Agent to draft your Annex IV Technical Documentation.
🚀 Quick Start
1. Install
pip install ai-act-check
2. The Smoke Test (No API Key needed)
Scan your repo for "High Risk" dependencies locally.
ai-act-check scan ./my-python-project
Output: JSON file with detected libraries and risk triggers.
3. Generate the Legal Draft (Requires API Key)
Turn your code scan into a formal legal draft.
export OPENROUTER_API_KEY="sk-..."
ai-act-check draft scan_results.json
Output: ANNEX_IV_DRAFT.txt written in formal legal prose.
📂 How it Works
- AST Analysis: We don't just look at
requirements.txt. We parse your actual code to see how libraries are used. - Risk Mapping: Deterministic mapping of libraries to Annex III use cases (Biometrics, Critical Infra, Employment).
- Agentic Drafting: An LLM agent (Compl-AI) acts as a Junior Associate to draft the text, strictly grounded in the scan evidence.
⚠️ Disclaimer
This tool is for technical documentation assistance only. It is not legal advice. You remain responsible for your compliance. For full conformity assessments and liability protection, visit Sovereign Code.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file ai_act_check-0.1.0.tar.gz.
File metadata
- Download URL: ai_act_check-0.1.0.tar.gz
- Upload date:
- Size: 11.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
a49e3f9d1dc8f1df6133e65cc7741a740b52ad2c27b8009f5e6b414d16b5bedf
|
|
| MD5 |
389ff88e97936c1455d605703194510c
|
|
| BLAKE2b-256 |
25247ffa54e057888c47b922c8e1cf33c8b778f74e539e2a60301dcdc6b39952
|
File details
Details for the file ai_act_check-0.1.0-py3-none-any.whl.
File metadata
- Download URL: ai_act_check-0.1.0-py3-none-any.whl
- Upload date:
- Size: 11.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
40c0a6c09c582da19c99a7d1e4900d4a76dd61545a00aab7fdc7c31e6aaf4f97
|
|
| MD5 |
0d35e3e830ade0147510b4f9cc183322
|
|
| BLAKE2b-256 |
9f5ff6b1245a50f6f069f33290dbf855c759462098b7bfb09b3fad760dae86b7
|