Skip to content

MCP Leaderboard

SMB Compare leads our MCP board with 100 of 100; we check 8,211 servers every day

·
#1 by our score, 100 of 100
SMB Compare
MCP servers checked daily
8,211
open to any client, no sign-in
54%

MCP Leaderboard

CSV · 40 rows

In the week of October 19 – 25, 2026, the Atlas checked 8,211 MCP servers every day and 8,169 answered. Of those, 54% are open to any client; the rest want a sign-in. Of the 10 most used servers in the Atlas that answered, 10 answered every daily check. The median open server lists 7 tools. Top of the board: SMB Compare, score 100 of 100, answered 7 of 7 daily checks. 444 servers have a full record: a week of checks, tools read, use cases judged. 27 score a perfect 100. This is the first week of daily checks, so nothing is compared with a week before yet.

Top by our score

Top by our score · Oct 19 – 25, 2026
Server#ScoreAnswered this weekToolsSign-inAdvice
SMB Compare11007 of 72openconnect
CheckMyVIN21007 of 72openconnect
Cocktail Glass31007 of 77openconnect
mailX by mailwarm41007 of 717openconnect
Fromerica51007 of 78openconnect
Epoch Calculator61007 of 78openconnect
Hesper Atlas71007 of 715openconnect
Cron Translator81006 of 63openconnect
Can I enter?91006 of 63openconnect
Harcourt Valuations101005 of 52openconnect
Erik Bethke111006 of 610openconnect
Marketcaper121005 of 54openconnect
SteamGPT131007 of 79openconnect
AviationHS141005 of 55openconnect
Remove AI Meta151005 of 52openconnect
Sky Valley161005 of 55openconnect
We Ultrarich171005 of 55openconnect
Energy Podcast Digest181005 of 55openconnect
Semscraper191005 of 58openconnect
Eagle Virtual201005 of 56openconnect

Most used

Most used · Oct 19 – 25, 2026
Server#ScoreAnswered this weekToolsSign-in
NVIDIA11007 of 7–OAuth
Figma2967 of 7–OAuth
Wolfram3867 of 73open
PayPal41007 of 7–OAuth
ZoomInfo51007 of 7–OAuth
Zapier6937 of 7–OAuth
Cloudflare7737 of 72open
CoinMarketCap81007 of 7–OAuth
RocketReach9787 of 7–OAuth
Atly10857 of 77open

Who moved

None this week.

Did not answer our checks

Did not answer our checks · Oct 19 – 25, 2026
ServerAnswered this weekUptime the week before
Papaya3 of 7–
Ascent Partners3 of 7–
Daisycon, part of Linehub3 of 7–
RaiseKit3 of 7–
Talarion3 of 7–
ResumeAI2 of 7–
PROME1 of 7–
Bavlio2 of 6–
Address to ZIP2 of 6–
YantraKit2 of 6–

Does your site have an entrance for AI agents? Free check →

Method

The INITE Atlas asks every vendor's own MCP server once a day whether it answers, openly or asking for a sign-in, and how fast; an open server's tools are read and graded. The score is the Atlas's own, out of 100: reliability over 30 days, the share of the product's use cases the tools do, how well the tools are defined, what they cost an agent's context, public adoption and how readily a client gets in.

The first board is the servers people know: the Atlas's most used vendors whose server answered at least once this week, in order of use, with their score where they have one. Most of them sit behind a sign-in, so they have no full record and are not on the score board.

The board ranks only servers with a full record: at least a week of checks, tools read, use cases judged. A server behind a sign-in shows nobody its tools, so it is counted in the week's numbers but not ranked. Ties are broken by the week's checks, then by how much the vendor is used in the Atlas.

A mover's score is compared with the score frozen the week before, and a change under 10 points is left out; its uptime is compared with the week before's checks, and a change under 30 points is left out. A server is listed as not answering our checks when fewer than half of at least four checks got an answer after it had been up. That is what our checker saw, not a verdict on the vendor: a firewall, a regional block or a rate limit on our side looks the same. Nothing here is written by a model: the sentences are filled in from the numbers in the tables.

How the score is counted, with its weights

Free to cite with a link to this page (CC BY 4.0).

Browse every MCP server in the Atlas