Index / Verified tools

Verified MCP tools, by capability

Every other registry indexes servers: a name, a URL and whatever prose the publisher wrote. The tool names and input schemas an agent actually has to match on exist in no index. So we ask each server directly, and store what it answers.

Files and storage
1,162 tools · 433 servers
Analytics and monitoring
1,131 tools · 400 servers
Calendar and scheduling
739 tools · 293 servers
Payments and billing
725 tools · 289 servers
Crypto and blockchain
725 tools · 177 servers
Images, audio and video
658 tools · 182 servers
Maps, location and weather
612 tools · 299 servers
Web scraping and browser automation
519 tools · 91 servers
PDF and document extraction
486 tools · 161 servers
Social media
481 tools · 69 servers
E-commerce
441 tools · 152 servers
HR and recruiting
437 tools · 151 servers
Finance and market data
416 tools · 119 servers
Email
355 tools · 134 servers
Science and research
316 tools · 135 servers
GitHub and version control
275 tools · 130 servers
Legal and government data
234 tools · 91 servers
CRM and sales
193 tools · 93 servers
AI models and embeddings
185 tools · 114 servers
Security and compliance
184 tools · 72 servers
Messaging and chat
143 tools · 40 servers
Cloud and DevOps
126 tools · 47 servers
Databases and SQL
107 tools · 41 servers
Web search
84 tools · 48 servers
Project management
70 tools · 20 servers

What is in the index

Measured, not claimed. This is the whole index by kind, 11,380 resources in total.

MCP servers8,291
Agent skills2,388
Machine-readable docs264
Other callable resources180
OpenAPI services153
A2A agents63
Catalogues15
ARD registries12
Packages6
Plugin manifests3
GraphQL APIs2
Agent descriptors (ACP, OASF, AgentFacts)1
Datasets1
WebMCP browser tools1

How the ecosystem spells the same thing

Media types as publishers actually wrote them. Filters match exactly, so a registry that does not normalise these silently drops thousands of entries. We normalise all of them, plus both URN prefixes in circulation (urn:air: and urn:ai:), on ingest.

application/mcp-server-card+json4,194
application/mcp-server3,706
application/ai-skill1,456
application/ai-skill+md789
application/mcp-server+json386
application/json144
text/markdown107
application/vnd.github.copilot-plugin102
application/openapi+json85
text/html84
application/a2a-agent-card+json60
application/vnd.oai.openapi+json51
This is not a directory of every server. We measured what ranks for MCP queries and the primary source wins: mirroring it at scale adds nothing. These pages exist because the verified tool surface is data that is not published anywhere else.

The whole corpus is also an open dataset under CC BY 4.0: huggingface.co/datasets/AgenticResourceDiscovery/verified-mcp-tools.