GitHub agent finder is now available. Instead of hand-wiring which MCP servers, skills, canvases, agents, and tools each agent should use—and filling your context window in the process—GitHub Copilot can now discover the right capability for a task on its own.

Describe a task in plain language and agent finder searches an index of available AI resources, then returns ranked matches Copilot can pull in on demand from any catalog you point it at. Agents load what the work calls for instead of carrying every tool just in case, and the best options are ranked for you to help you decide what to install.

Agent finder implements the open Agentic Resource Discovery (ARD) specification, developed in collaboration with Google, GoDaddy, Hugging Face, and Microsoft. It launches alongside Microsoft’s announcement of the ARD specification, which features GitHub.

Agent finder works against a registry, and you decide which one:

• Discover from a registry you choose: Point agent finder at GitHub’s curated public catalog, or at your own private registry of internal resources, and the answers an agent gets are scoped to that source.
• Enforce with managed settings: The same place you govern Copilot is where you define which resources agents are allowed to discover and use. Agent finder only ever surfaces what your enterprise permits.
• No auto installation: Agent finder finds the right tool at the right time. It doesn’t silently connect anything. You stay in control of what actually gets wired in.

Agent finder is available on all GitHub Copilot plans. Because it implements the open ARD specification, any registry or AI client can adopt the same discovery model.

Try agent finder to discover the right capabilities for your tasks.