Agentic Resource Discovery proposes a common way for agents to search for capabilities when a task needs them. As Hugging Face explains, a publisher describes its resources in a catalog, and a registry returns ranked matches to a natural-language query. Discovery happens before the AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → connects to a tool or loads instructions.
Hugging Face's implementation searches hosted applications and skills, filters application results to running services, and returns the resource format the client requests. This gives developers a route beyond manually configuring every capability in advance. The specification was presented as a draft, with additional federation support still planned.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
articleAgent Papersagentpapers.dev
articleCartograph: Federated Tool Discovery with Operator-Attested Retrieval for AI AgentsJustice Owusu Agyemang, Michael Agyare, Kwame Opuni-Boachie Obour Agyekum, Kwame Agyeman-Prempeh Agyekum, Francisca Adoma Acheampong, Jerry John Kponyo