[AI-interop] /context endpoint
Tim Jenness
tim.jenness at noirlab.edu
Sun Nov 8 05:11:36 CET 2026
On Nov 8, 2026 at 12:00:59, Dave Morris via AI-interop <ai-interop at ivoa.net>
wrote:
> Hiya,
>
> Following on from the apps session on AI adoption. I had a quick
> conversation with Google AI chat (gemini-3.8-flash) and together we came
> up with the following suggestion:
>
> ...
>
> General idea is the /context endpoint is a list resources an agent
> should read before trying to use the service and the data it provides.
> It allows data providers to curate the instructions for their own
> particular resource.
>
>
For SIA I think this would be a pretty simple thing to have because I think
SIA already has a mechanism where you can ask for a VOTable that returns
the ranges of acceptable parameters and distinct enum-like values. So we
already have code in the service that is summarizing things.
I still think that something like this should be augmented by improving
semantic linkages of result columns within the results (at Rubin we return
multiple fluxes for each filter but there is no easy way to link ugrizy
together per algorithm or point at an algorithm reliably). A /context
endpoint with information that can not be determined from the models is
good but it’s even better if the results have more information.
—
Tim Jenness
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ivoa.net/pipermail/ai-interop/attachments/20261107/1b6944a4/attachment.htm>
More information about the AI-interop
mailing list