el1ght

el1ght / aulite

Public

EU AI Act compliance proxy for AI systems. Drop-in HTTP proxy that monitors every AI interaction for regulatory risks, logs to a tamper-proof audit trail, and generates legal-grade PDF reports. 143 rules across 8 Annex III domains. Self-hosted, open-core

10
0
100% credibility
Found Mar 26, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Aulite is a self-hosted tool that intercepts AI requests to detect compliance risks under the EU AI Act, maintains secure logs, provides a dashboard, and generates regulatory reports.

How It Works

1
๐Ÿ” Discover Aulite

You learn about a helpful tool that watches your AI chats to keep them legal under new EU rules.

2
๐Ÿš€ Launch easily

Start it with one simple command, and your compliance helper runs right on your computer.

3
๐Ÿ”— Link your app

Update one web address in your AI app so all its thinking goes through the helper first.

4
๐Ÿ›ก๏ธ Watch it protect

Every question and answer gets checked instantly for risks, with safe logs building automatically.

5
๐Ÿ“Š Check the dashboard

Open a friendly web page to see live stats, flagged issues, and easy overviews.

6
๐Ÿ“„ Make reports

Click to create polished PDF documents proving your setup meets the rules.

โœ… Proven safe

Relax knowing you have tamper-proof records and reports ready for any review.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is aulite?

Aulite is a self-hosted TypeScript HTTP proxy that slots between your app and AI providers like OpenAI or Anthropic, scanning every request and response for EU AI Act violations across 143 rules in 8 high-risk domains like HR and finance. It logs interactions to a tamper-proof SQLite audit trail with SHA-256 hash chaining, serves a React dashboard for real-time stats, and spits out legal-grade PDF reports for audits or incidents. Just swap your client's base_url to localhost:3000 and docker run itโ€”no code changes needed.

Why is it gaining traction?

With EU AI Act enforcement hitting August 2026 and fines up to 7% of global revenue, devs need an ai act compliance tool that works out-of-the-box without slowing down streaming APIs (<5ms overhead). Unlike basic checklists, it auto-detects PII, prohibited practices, and discrimination in real-time, plus optional LLM judging and one-click act compliance certificates via PDF exports. The CLI for init/start/domains and Docker Compose setup make it dead simple for local testing.

Who should use this?

EU-based engineering leads building AI for HR screening, credit scoring, or biometrics, where Annex III rules apply. Compliance officers at fintechs or edtechs wanting an ai act compliance checker to proxy production traffic and generate FRIA drafts. Teams self-hosting OpenAI-compatible endpoints like Ollama who need data act compliance without vendor lock-in.

Verdict

Solid prototype for ai act complianceโ€”10 stars and 1.0% credibility reflect early days, but 108 passing tests, polished docs, and Docker image show promise. Grab it now if you're in a high-risk domain; it'll mature fast as fines loom.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.