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.

Analytics and monitoring
3,514 tools · 845 servers
Files and storage
3,506 tools · 938 servers
Maps, location and weather
3,367 tools · 726 servers
Payments and billing
2,436 tools · 601 servers
GitHub and version control
2,373 tools · 229 servers
Calendar and scheduling
2,212 tools · 640 servers
Crypto and blockchain
1,968 tools · 450 servers
Images, audio and video
1,874 tools · 356 servers
HR and recruiting
1,432 tools · 327 servers
Cloud and DevOps
1,431 tools · 196 servers
E-commerce
1,249 tools · 280 servers
PDF and document extraction
1,187 tools · 285 servers
Science and research
1,165 tools · 257 servers
Databases and SQL
862 tools · 141 servers
Email
778 tools · 242 servers
Finance and market data
687 tools · 206 servers
Web scraping and browser automation
641 tools · 140 servers
Social media
632 tools · 85 servers
Legal and government data
356 tools · 117 servers
CRM and sales
352 tools · 124 servers
Security and compliance
311 tools · 106 servers
AI models and embeddings
292 tools · 177 servers
Messaging and chat
267 tools · 72 servers
Web search
232 tools · 105 servers
Translation and language
146 tools · 50 servers
Project management
141 tools · 35 servers

What is in the index

Measured, not claimed. This is the whole index by kind, 15,882 resources in total.

MCP servers9,323
Agent skills3,076
OpenAPI services2,770
Machine-readable docs322
Other callable resources253
A2A agents96
Catalogues16
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+json5,201
application/mcp-server3,679
application/vnd.oai.openapi+json2,575
application/ai-skill1,938
application/ai-skill+md976
application/mcp-server+json438
application/json172
application/openapi+json170
text/markdown132
application/vnd.github.copilot-plugin104
text/html98
application/a2a-agent-card+json82
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.