Agentic Resource Discovery · now an IETF Internet-Draft

Every tool an agent
can call, in one index.

Neuronto is a federated ARD registry. Ask it what can do a task and it searches its own catalogue and every other public ARD registry at once, then fuses the rankings into a single answer. It currently indexes 194,370 resources from 108,500 publishers, 188,282 verified to respond.

What is ARD?

ARD stands for Agentic Resource Discovery. It is an open specification for how AI agents find the tools, skills, agents and APIs they need, published in June 2026 by a working group including Google, Microsoft, Hugging Face, AWS, Cisco, GitHub, Nvidia, Salesforce and Snowflake.

An agentic resource is anything an AI client can call to get work done: an MCP server, an A2A agent, a skill, an API, a workflow. ARD answers one question, “what is available for this task?”, and then gets out of the way. It is a discovery layer, not a runtime, and it does not replace MCP or A2A: it tells an agent what exists, and the agent connects using the resource's own protocol.

Why agents need it

An agent can only use capabilities someone wired in beforehand, and every tool description competes for space in the context window. That does not survive an ecosystem of thousands of tools. ARD moves selection out of the context window and into a search service.

How discovery works

Publishers describe what they offer at /.well-known/ard.json on their own domain. Registries crawl those manifests and answer queries over them. Nobody has to be approved by a marketplace to be found.

What federation means

The specification defines three federation modes and makes auto the default: a registry queries its peers, merges their results and returns one set. Ask once, reach the whole federation.

Who it is for

Agent builders who want to stop hard-coding integrations. API and tool vendors who want to be discoverable without applying anywhere. Platform teams running discovery over their own internal services.

The index

Resources indexed
194,370
Publishers
108,500
Verified live
188,282
Registries federated
6

What is in the index

194,370 entries · 108,500 publishers

Every entry is a resource an AI client can actually call, held with its endpoint, its description, the queries its publisher says it serves, and which sources carry it. The whole index is queryable over the API with no key and no signup, and browsable page by page through GET /agents.

Kind of resourceEntriesWhat it is
mcp-server176,225MCP servers an agent can connect to and call as tools
skill12,278Skills and agent skill bundles
openapi3,014REST APIs described by an OpenAPI document
other914Resources whose type is outside the named set
doc886Machine-readable documentation such as llms.txt
plugin592Plugin manifests, including the OpenAI-era ai-plugin.json
a2a-agent333A2A agents with a published agent card
catalog97Nested catalogues pointing at further entries
registry18Other ARD registries, which is how federation is discovered
package7Published packages
graphql2GraphQL APIs
webmcp2Tools a web page exposes to a browser agent, W3C WebMCP
agent1Agent descriptors that are not A2A cards: ACP, OASF, AgentFacts
dataset1Published datasets

Where it comes from

Aggregated from every public ARD registry, the official MCP Registry, the major community aggregators, and our own crawl of the well-known discovery paths. No single source is close to complete on its own: the official registry does not carry servers their authors never submitted to it, and each aggregator holds thousands the others do not, so the table below is the union rather than any one of them. The crawl reaches publishers no other index covers: the only other general crawler in the ecosystem stops at the top 100,000 domains, so anyone smaller is invisible everywhere else.

SourceEntries
glama74,913
Official MCP Registry67,606
mcpso18,482
smithery10,239
GitHub Agent Finder9,347
Our own crawl of the discovery paths5,438
WellKnown5,168
Hugging Face Discover3,315
apis.guru2,509
ardregistry1,684
submitted967
Desvela869
dockerhub100

Publishers

PublisherEntriesVerified liveKinds
mcp.so
mcp-server
18,48218,4811
github.com
mcp-server · openapi · plugin
8,5136,6123
huggingface.co
mcp-server · registry · skill
3,2933,1303
io.github.sadri-dridi
mcp-server
2,3662,3661
io.github.mrfentmen
mcp-server
2,0552,0541
pipeworx.io
mcp-server
2,0482,0481
io.github.pipeworx-io
mcp-server
1,9371,9371
ora.ai
a2a-agent · mcp-server · skill
1,6661,6563
pipeworx-io
mcp-server
1,5391,5231
io.github.mcp-dir
mcp-server
1,1201,1201
mcp-dir
mcp-server
1,1191,1191
io.usefulapi
mcp-server
1,0451,0361

Every public ARD registry, measured

Benchmarked 31 August 2026 with the specification's own conformance tool. Precision is top-1 over six queries with unambiguous answers; spread is the average gap between the best and worst score in a result set.

RegistryConformanceEntriesP@1 SpreadLatencyfederation: auto
NeurontoPASS194,370 6/678<200msyes
GitHub Agent FinderFAIL3446/6 32.51,970msno
WellKnownPASS1,2396/6 6.7401msno
Hugging Face DiscoverPASS4120/6 40.81,712msno
DesvelaPASS932/6 39.7213msno

Use it

ARD v0.91 + IETF drafts · no key, no signup

Search the federation

curl -s https://neuronto.com/search \
  -H 'content-type: application/json' \
  -d '{"query":{"text":"scrape a website"},
       "federation":"auto"}'

Install as an MCP server

claude mcp add --transport http \
  neuronto https://neuronto.com/mcp

Publishing? Serve an ARD manifest at /.well-known/ard.json with representativeQueries on every entry. The crawler finds it. No submission form, no allowlist.