Agentic Resource Discovery Specification
Posted by damick 1 day ago
Comments
Comment by gavinray 15 hours ago
The best solution we could find for this problem was a combination of a Wiki + "Saved Programs" (executable scripts).
When you ask a question, the system does a fuzzy Wiki search to see if any topic has relevant info. Wiki links serve as "graph edges" that form a Knowledge Graph.
So lets say I ask "What products from the FOOBAR API sold least last month?"
The agent would look up "FOOBAR API", and then write a script to call the "products" endpoint with a date range + SQL fetch from "sales" DB and do whatever it needs to do.
If none of this info exists (URL to FOOBAR, location of "sales" data), the agent asks for more info and offers to update the wiki for future.
Comment by xtiansimon 33 minutes ago
Comment by nimonian 14 hours ago
Comment by gavinray 15 hours ago
Too many protocols got created (MCP, A2A, etc) and so now you've got to create a new standard that can consume them all, and Google being a search index would like to index them, please =)
[0]: https://modelcontextprotocol.io/specification/2025-11-25/ser...
Comment by Catloafdev 15 hours ago
Comment by nostrebored 8 hours ago
Imposing a structure on this unstructured discovery feels like it ignores all progress in IR, and imposes a farcical structure that doesn’t have a tangible benefit.
Comment by myaccountonhn 1 hour ago
Comment by _pdp_ 13 hours ago
There is no need for another file IMHO.
Comment by iandanforth 15 hours ago
Agents can use Google. They can also see all the same signals humans do to judge quality. Maybe there's a need for a specialized directory, but does it need a spec or an acronym? No.
Comment by ivanbelenky 11 hours ago
Comment by khaledh 11 hours ago
Comment by wslh 16 hours ago
Comment by verdverm 12 hours ago
Comment by 10cmd 14 hours ago