Cleo Labs
Automate global compliance for selling physical products
Start new thread

Cleo Atlas Legal API - The world's regulations, machine-readable.

by
•
Cleo Atlas turns 256,000+ regulations across 177 jurisdictions into one machine-readable API. Two atlases: Product (cosmetics, devices, food, toys, electronics) and Global (finance, AI, data, ESG). Query any law, anywhere. Built by Cleo Labs.

Add a comment

Replies

Best
Maker
📌
Hi Product Hunt, Naomie here, co-founder at Cleo Labs! Two years ago we started Cleo because we kept seeing the same painful scene: a brand spends nine months building a beautiful new product, then learns three weeks before launch that one ingredient is banned in the country they were betting on. Or that the EU AI Act actually applies to that little internal model they shipped last quarter. Every time, the gap was the same — the law existed, it was public, but it was unqueryable. So we built the layer underneath. Cleo Atlas is one API exposing two atlases: a Product Atlas with 46,031 regulations across 20 product categories and 50 jurisdictions, and a Global Atlas with 210,508 regulations across 177 jurisdictions — health, finance, environment, AI, data, labor, tax. 234M+ documents normalized into clean JSON by MARIA, our retrieval engine. Today it already runs inside the biggest brands. We are now opening it so anyone building a compliance product, a legal copilot, or an AI-governance layer can stop scraping ministry websites at 2 a.m. The explorer is free, no credit card: legaldata-public.cleolabs.co. Hit it, break it, ask it weird questions. I would love your honest feedback especially: which regulation would you query first, and what does our API miss for your use case? I will be in the thread all day.

Well good luck rank is already yours even if you feature it for monthly 🙌🎉

Maker
thank you for your continuous trust !!

curious about the underlying data model. how granular is the normalization process? can developers access both raw source references and structured outputs?


Maker

 We normalize at the obligation level, not the document level each regulation is broken into discrete requirements with their own metadata (jurisdiction, authority, product category, effective dates). That's what lets us answer "what applies to this product in this market."

really interesting launch. how do you handle multilingual regulations and legal terminology? is everything normalized into a single schema?

Maker

 Unified schema, but original source preserved plus an interpretation layer, and ambiguous terms go to human validation rather than being flattened :)

Maker
🚀🚀🚀🚀🏆🏆
Maker

 thanks for contributing a lot on this one!!!

very interesting approach to regulatory intelligence.how do you handle conflicting regulations across multiple jurisdictions? does the API surface those conflicts automatically?


Maker
Great question! Yes, it’s core: when a product hits diverging requirements across markets, the API surfaces the conflict expli

Congrats on the launch!
The Digital Product Passport deadline in 2030 is real and most brands are completely unprepared. Is that already part of what Cleo maps for EU markets?

Maker
Thanks! And yes, DPP is squarely in scope. We map ESPR / Digital Product Passport requirements for the EU and you’re right, most brands are nowhere near rea

The human-in-the-loop validation piece is what makes this credible to me.
How long does a full compliance map take from URL submission to verified output?

Maker
thanks Abdul! The pipeline runs in minutes, the rest is human validation depending on jurisdiction complexity. Speed where it’s safe, human eyes where it counts. :)

interesting product! can users build jurisdiction specific compliance dashboards on top of the API? or is Atlas primarily focused on data retrieval?


Maker
Thanks Luz! Both 🙂 Atlas is the structured open-data layer, the API lets you build your own jurisdiction-specific dashboards on top

Regulatory data is notoriously messy. how does MARIA deal with ambiguous or poorly formatted government documents? are confidence scores exposed through the API?