Grammar might be weird because I'm not familiar with English.

rawctx is a registry of semantic models, and it's a service that goes against this in a time when AI products are full. It's not using AI, it's a service that helps AI.

think npm, but for semantic models

Several companies recently agreed on a common format called Open Semantic Interchange (OSI) and released v1.0. rawctx is built to be the independent registry that no single vendor can own.

The reason I created this is that the definition of semantic models is still ongoing, and I've seen the potential in that it's a format that can be shared.

Currently, I'm working as a data engineer, modeling semantic layers at the company I'm working for. And I found out in this process that I need to share models within my team.

However, it was difficult to measure the performance of each service on the same semantic model basis because the team was using various programs on a trial basis.

So I thought, let's make a service where we can share this.

That's why I made it. I have a few semantic models shared on a trial basis now, but I hope it will be a service that many people can use comfortably over time.

In addition, since the semantic model is an important asset for a company, we have also prepared a space for it to be shared privately.

install : pip install rawctx

commands :

rawctx search "shopify revenue"

rawctx info @dbt-hub/jaffle-shop

rawctx snapshot-download @pasar6987/sf-revenue-asset << downloadable

GitHub administrators of indexed models can transfer ownership after login through rawctx claim.

rawctx claim @github-hub/osi