Velocity Labs is considering taking over the maintenance of the XCM Global Registry. We have already asked a few teams building on Polkadot about their reliance on the project and what improvements everyone is looking for.
From what we gathered, only a few projects are actually using it while most report seeing the value but either just don’t need it or would need some technical improvements for it to become a viable dependency (namely, fixing some wrong multilocations, improving the data structure of the tokens’ metadata and better indexing by multilocation, by local asset id, etc).
Do you use it? What for? What’s lacking or not working at the moment? What other improvements would you be keen on?
I was exploring the XCM Global Asset Registry, but as you pointed out the need of technical improvements. I found some issues like missing assets from Polkadot Asset Hub list and inconsistent formatting.
Would love to have a production ready XCM registry
As you mention, the AT API registry is just a wrapper around the GAR so it’s not quite a competitor. Regardless, I agree we should thrive to maintain an excellent one rather than keeping two that are not delivering on what real users and teams need.
What features do you miss from the AT API Registry? And what’s been helpful for your project?
As mentioned, the asset-transfer-api-registry utilizes the xcm-global-registry as part of the asset registry it provides, pulling in its xcAssets data daily. This data is used within the Asset Transfer API for creating XCM transactions from parachains.
The two notable things that have been issues recently for us are the parents values for global consensus Polkadot assets having incorrect values and junctions with multiple keys (e.g. GeneralKey junctions) appearing as nested arrays within asset locations. Both of these issues have been fixed but improvements to the xcm-global-registry’s stability/usability are welcomed.
