πŸ‘‘

MCP QUEEN

The Graded Registry

Every remote server in the official MCP registry, probed live and graded with receipts β€” the only MCP index where every grade shows its evidence.

16246servers indexed
7735remote endpoints
742graded (rolling)
206grade A
68%reachable
384msmedian latency
SortBestWorstJust probedMost toolsFastest
GradeAllA (206)B (3)C (194)D (105)F (234)Provisional
CategoryAll categoriesOther (202)Web & Search (136)AI & Agents (127)Cloud & Infra (65)Productivity (42)Dev & Code (34)Data & Databases (28)Finance & Crypto (26)Communication (21)Media & Design (19)Commerce (17)Science & Health (13)Security (12)

21 matches. Categories are keyword-derived from registry metadata β€” imperfect by design, deterministic by principle.

#ServerGradeScoreLatencyToolsAuthProbed
1 ai.trendsmcp/amazon
Amazon Trends MCP
Web & Search
D 59
954ms β€” open 07-12 18:02
2 ai.trendsmcp/google-trends
Google Trends MCP
Web & Search
D 59
863ms β€” open 07-12 18:02
3 ai.trendsmcp/youtube
YouTube Trends MCP
Web & Search
D 59
859ms β€” open 07-12 18:16
4 ac.inference.sh/mcp
inference.sh
Web & Search
D prov. 57
564ms β€” auth-wellbehaved 07-12 14:18
5 ai.docuwriter/docuwriter
Web & Search
D prov. 57
723ms β€” auth-wellbehaved 07-12 14:19
6 ai.findip/patent-search
FindIP
Web & Search
D prov. 57
793ms β€” auth-wellbehaved 07-12 14:31
7 ai.openhelm/openhelm
Web & Search
D prov. 57
512ms β€” auth-wellbehaved 07-12 15:16
8 ai.tinyfish.agent/web-agent
TinyFish AI Web Agent
Web & Search
D prov. 57
678ms β€” auth-wellbehaved 07-12 18:02
9 ai.trycanonical/company-search
Canonical Company Search
Web & Search
D prov. 57
745ms β€” auth-wellbehaved 07-12 18:16
10 ai.verant/verant
Verant
Web & Search
D prov. 57
505ms β€” auth-wellbehaved 07-12 18:16
11 ai.websitepublisher/mcp
Web & Search
D prov. 57
2055ms β€” auth-wellbehaved 07-12 18:33
12 app.newsmind/mcp
Newsmind
Web & Search
D prov. 57
669ms β€” auth-wellbehaved 07-12 19:03
13 ai.parallel/search-mcp
Parallel Search MCP
Web & Search
D prov. 54
156ms β€” auth-bare 07-12 15:16
14 app.aspirelearning/mcp
Web & Search
D prov. 54
521ms β€” auth-wellbehaved 07-12 18:33
15 app.govauctions/govauctions
Web & Search
D prov. 54
24ms β€” auth-bare 07-12 18:48
16 ai.mino/web-agent
TinyFish AI Web Agent
Web & Search
D prov. 52
1637ms β€” auth-wellbehaved 07-12 15:01
17 ai.1325/mcp
1325.AI
Web & Search
D prov. 51
1087ms β€” auth-wellbehaved 07-12 14:18
18 ai.deinai/creator-skill
Creator SKILL β€” Influencer Discovery
Web & Search
D prov. 49
1100ms β€” auth-bare 07-12 14:19
19 ai.deinai/creator-skill-v2
Creator SKILL v2 β€” Influencer Discovery
Web & Search
D prov. 49
1039ms β€” auth-bare 07-12 14:19
20 ai.moxt/mcp-workspace
Moxt Workspace
Web & Search
D prov. 49
345ms β€” auth-bare 07-12 15:01
21 app.opendealer/mcp
OpenDealer MCP Server
Web & Search
D prov. 45
752ms β€” auth-bare 07-12 19:03

Methodology

Each server is probed live over streamable HTTP: reachability (25) β€” does initialize succeed; auth-gated servers earn partial credit only if they advertise WWW-Authenticate so clients can discover OAuth. protocol (15) β€” valid JSON-RPC handshake; deprecated SSE transport is penalized. tooling (35) β€” tools/list works, share of tools with descriptions and fully-typed input schemas, median description depth. latency (10) β€” initialize round-trip. provenance (15) β€” registry metadata completeness and whether the reverse-DNS namespace actually matches the serving domain. Scores scale to what is verifiable; unverifiable dimensions mark the grade provisional rather than guessing. Every point carries the verbatim observation that earned it. No stars, no votes, no pay-to-rank β€” probes only.