Best MCP Servers: The Complete Ranked List

TL;DR: This best MCP servers list ranks 163 Model Context Protocol servers from the official MCP Registry on real adoption data. Top of the list on our composite quality score is Codebase Memory at 96/100; the most-downloaded is Chrome DevTools MCP at 1,563,379 weekly package downloads. 92 of 163 (56%) are first-party servers built by the vendor whose product they connect to, and every one carries copy-paste install snippets for 4 AI clients.

Last updated: 7 August 2026. Data pulled: 7 August 2026, from the official MCP Registry plus the GitHub API, npm, PyPI and the GitHub Advisory Database. Servers ranked: 163. Combined weekly package downloads: 2,256,524.

Data note. GitHub enrichment covers 161 of the 163 servers in this build, so star counts, licences and release history are reported directly from the GitHub API.

An MCP server is a program that exposes tools, data and prompts to AI applications through the Model Context Protocol, the open standard Anthropic released in November 2024. It is the adapter that lets Claude, Cursor, VS Code or ChatGPT read from and act on an external system without bespoke integration code.

Most MCP lists are a copy of somebody’s GitHub “awesome” file. This one is built from the official registry and enriched with adoption, maintenance and security data, so you can compare servers rather than just discover them.

The Numbers Worth Quoting

• 163 MCP servers ranked as of 7 August 2026 across 11 categories. 44 of them (27%) ship an MCP-specific npm or PyPI package, together drawing 2,256,524 downloads a week.
• Chrome DevTools MCP is the most downloaded MCP server at 1,563,379 a week, 69.3% of attributable download volume. The top three take 92.5% and the top ten 97.8%.
• 26 servers are excluded from those download totals because their package is a wider library rather than the MCP server, setting aside 13,655,940 weekly downloads (86% of the raw sum). Most MCP rankings sum these in.
• 92 of 163 (56%) are first-party, published under a registry-verified namespace the vendor owns. They account for 98.9% of download volume.
• Developer Tools is the biggest category at 54 servers (33%), followed by Other at 26.
• 116 of 163 run over stdio as a local process. Only 62 (38%) offer Streamable HTTP, and just 60 publish a hosted remote endpoint.
• All 163 servers ship install snippets for Claude Desktop, Cursor, VS Code and Claude Code, so adding any of them is a copy-paste.
• 78 publish to npm and 34 to PyPI, which is why Node and Python are effectively the two runtimes of the MCP ecosystem.
• 6 security advisories across the whole list, and every one of them is already patched. 108 servers (66%) have no known advisory, 5 carry a patched one, none carry an unpatched one, and 50 cannot be checked at all because they publish no queryable package.
• Maintenance is strong: median 3 days since the last update, 142 servers (87%) updated within 30 days, and none untouched for over a year.
• Nothing on this list is flagged deprecated in the registry. Deprecated servers are not hidden when they do appear, because one still turns up in search results elsewhere and you deserve to know its status.

Every figure is reproducible from the tables below. Methodology and citation line at the bottom.

What Is an MCP Server, and What Does This List Cover?

An MCP server is a small program that wraps a system, a database, a GitHub repo, a Slack workspace, a browser, and exposes it to an AI client through three primitives: tools the AI can call, resources it can read, and prompts it can reuse. The client talks to it over JSON-RPC, either locally over stdio or remotely over Streamable HTTP.

Inclusion here is automatic. A server is ranked when it appears in the official MCP Registry with a description and a working install method (an npm, PyPI, OCI or NuGet package, or a public remote URL), and clears an adoption floor. Servers marked deleted are excluded; servers marked deprecated are kept and badged, because you deserve to know a thing is deprecated rather than simply not find it.

One thing this list does that a GitHub “awesome” file cannot: it tells you whether the vendor built the server themselves. That distinction matters more than any score, and it gets its own section below.

The Most-Downloaded MCP Servers

There is no universal install counter for the Model Context Protocol, so weekly npm and PyPI download counts are the closest available proxy for a server actually being pulled into real projects. They measure installs rather than bookmarks, which is why they sit alongside stars in the adoption pillar rather than behind them.

Two caveats set the scope of this section, and the second one is the reason our numbers are lower than you will see elsewhere.

Downloads only exist for servers distributed as a package. A remote-only server behind a hosted URL, or one shipped solely as a container image, has nothing for npm or PyPI to count, so it is absent here rather than sitting at zero.

We do not count a download unless the package is the MCP server. Many registry entries point at a general-purpose library that merely also exposes an MCP server, and its download count measures the whole library. The largest single case is browser-use, whose registry entry points at `browser-use`: those 13,400,923 weekly downloads are people installing that library, the overwhelming majority of whom never touch its MCP server. Counting it would be the download equivalent of crediting a server with its parent project’s GitHub stars, which is why the engine already rejects borrowed stars and why we reject borrowed downloads on the same principle.

So 26 servers are excluded from every download total in this report, setting aside 13,655,940 weekly downloads, 86% of the raw sum. They appear in the full list with their download cell marked library-wide rather than as a number that is not comparable to the rest. What remains is 44 servers (27% of the list) whose package is the MCP server itself, drawing 2,256,524 downloads a week between them. Every download figure and percentage below is computed over that subset. If you see an MCP ranking where a single server holds the overwhelming majority of all downloads, it has almost certainly summed the libraries.

One further limit we cannot correct: a few vendors ship many servers through a single shared package, so that package’s downloads are attributed to the one server the registry attaches it to rather than split across the family.

Rank	MCP server	Category	Weekly downloads	First-party	Updated	Quality
1	Chrome DevTools MCP	Developer Tools	1,563,379	First-party	0d ago	93/100
2	Amazon ECS MCP Server	Cloud	408,361	First-party	179d ago	78/100
3	Azure MCP Server	Cloud	114,702	First-party	0d ago	89/100
4	ClickHouse	Databases	47,481	First-party	2d ago	88/100
5	Windows-MCP	Developer Tools	18,688	First-party	0d ago	89/100
6	SmartBear MCP	Other	14,179	First-party	0d ago	83/100
7	Power BI Modeling MCP Server	Other	14,080	First-party	7d ago	88/100
8	Deep Agentic Core MCP	Developer Tools	9,076	First-party	2d ago	79/100
9	CrowdStrike Falcon MCP Server	Developer Tools	8,764	First-party	1d ago	85/100
10	Zotero MCP	Search & Web	7,638	Community	2d ago	84/100
11	mcp	Databases	7,222	First-party	0d ago	81/100
12	Codebase Memory	AI & Memory	4,175	Community	0d ago	96/100
13	Rootly	DevOps & Monitoring	3,503	First-party	1d ago	82/100
14	ClaudeR - RStudio MCP Server	Developer Tools	3,275	Community	4d ago	76/100
15	OpenTakeoff	Developer Tools	3,260	First-party	0d ago	79/100

Read download counts for what they are even after that filtering. A server pulled into a CI pipeline or rebuilt into a container image every week racks up downloads fast, which is a genuine signal of production use but not the same as a headcount of humans. Chrome DevTools MCP at 1,563,379 a week is well clear of the rest of the field, and servers wired into automated environments get re-installed on every run.

The median packaged server gets 1,663 downloads a week. So the distance between the leader and the middle of the pack is several orders of magnitude, and most MCP servers are still early.

The Best MCP Servers by Quality Score

Rank	MCP server	Category	Quality score	Weekly downloads	First-party	Security
1	Codebase Memory	AI & Memory	96/100	4,175	Community	No known advisory
2	Repowise	Productivity	95/100	library-wide	First-party	No known advisory
3	World Monitor	Other	94/100	not reported	First-party	Not checked
4	Chrome DevTools MCP	Developer Tools	93/100	1,563,379	First-party	Advisory, patched
5	Agent Skills Search Server	Search & Web	93/100	not reported	First-party	Not checked
6	Scrapling MCP Server	Cloud	92/100	library-wide	Community	No known advisory
7	browser-use	Search & Web	91/100	library-wide	First-party	Advisory, patched
8	mcp-server	Productivity	91/100	not reported	First-party	Not checked
9	Oh My Posh Validator	Other	91/100	not reported	First-party	Not checked
10	XcodeBuildMCP	Productivity	90/100	not reported	First-party	No known advisory
11	apify-mcp-server	Other	90/100	not reported	First-party	Not checked
12	Azure MCP Server	Cloud	89/100	114,702	First-party	Advisory, patched
13	Windows-MCP	Developer Tools	89/100	18,688	First-party	Advisory, patched
14	Microsoft Fabric MCP Server	Other	89/100	882	First-party	No known advisory
15	sem	AI & Memory	89/100	not reported	First-party	No known advisory

This ranking weighs adoption alongside maintenance, growth, trust and security, so it surfaces servers that are well run rather than only widely installed.

Quality scores here run from 61 to 96 with a median of 77 and a mean of 78.2. That is a wider spread than the model or extension lists we publish, because MCP is young: the difference between a vendor-maintained server with a licence and a repo, and a weekend project published to the registry, is large and the score reflects it.

How Is the Quality Score Calculated?

Five weighted pillars: adoption 35%, maintenance 25%, growth 15%, trust 15%, security 10%.

• Adoption (35%) is a composite proxy, stated plainly. Adoption blends weekly package downloads with GitHub stars, both log-scaled. MCP has no universal usage metric, so this is never presented as an exact install count.
• Maintenance (25%) is how recently the server was updated, plus its release cadence.
• Growth (15%) is the 30-day trend, computed from our own stored snapshots, so it appears only once enough history exists. A server with no history shows no trend rather than a fabricated number.
• Trust (15%) is structural: a first-party namespace, an open-source licence, and a public repository.
• Security (10%) maps each server’s npm or PyPI package to the GitHub Advisory Database. A server with no queryable package, remote-only or container-only, shows “not checked” rather than a false all-clear.

No server pays to be listed or ranked, and there are no affiliate links anywhere on this page.

First-Party or Community? The Signal That Matters Most

92 of 163 servers (56%) are first-party: published under a reverse-DNS or GitHub-organisation namespace the registry verified as belonging to the vendor. The other 71 (44%) are community builds.

This is the single most decision-relevant fact in the dataset, and it is the one a star count cannot tell you.

An MCP server usually needs credentials. A database server wants a connection string, a Slack server wants a workspace token, a cloud server wants API keys with real permissions. When the server is first-party, you are trusting the same company you already trusted with the underlying product. When it is community, you are extending that trust to a third party, which may be entirely fine and may be better maintained, but it is a separate decision you should make knowingly.

First-party servers take 98.9% of all download volume, so the ecosystem is consolidating around vendor-built bridges faster than the raw server counts suggest.

Which Categories Have the Most MCP Servers?

Developer Tools leads with 54 servers (33% of the list), ahead of Other at 26. By download volume the leading category is Developer Tools, which takes 71.7% from just 54 servers.

Category	Servers	Weekly downloads	Share	First-party	Top server
Developer Tools	54	1,618,259	71.7%	21	Chrome DevTools MCP
Other	26	33,934	1.5%	23	World Monitor
Search & Web	24	12,794	0.6%	10	Agent Skills Search Server
AI & Memory	13	4,755	0.2%	6	Codebase Memory
Productivity	10	none attributable	n/a	8	Repowise
Communication	8	294	<0.1%	6	Atomic Mail
DevOps & Monitoring	8	3,604	0.2%	5	mockserver
Databases	7	57,911	2.6%	6	ClickHouse
Finance	5	none attributable	n/a	2	Finance Toolkit
Cloud	4	523,063	23.2%	3	Scrapling MCP Server
Design	4	1,910	0.1%	2	Figma-Context-MCP
All 11 categories	163	2,256,524	100%	92	Codebase Memory

That developer tools lead on count is unsurprising: the first people to wire an AI into a system are the people who build systems, and coding assistants were the first clients to ship MCP support. The categories to watch are the sparse ones. A category with two or three servers and real download volume is where the protocol is being adopted faster than it is being served.

How Do You Install an MCP Server?

Every server in this list ships copy-paste configuration for four clients, which is the practical reason to use this page over a plain directory.

AI client	Servers with a ready install snippet
Claude Desktop	163 of 163
Cursor	163 of 163
VS Code	163 of 163
Claude Code	163 of 163

An MCP server runs as a small program, locally over stdio or remotely over Streamable HTTP, that an MCP client connects to. To add one, open your client config and point it at the server: in Claude Desktop edit claude_desktop_config.json, in Cursor edit ~/.cursor/mcp.json, in VS Code add a .vscode/mcp.json, or run “claude mcp add” in Claude Code. Use the MCP Inspector to test a server before wiring it into your assistant.

The transport matters for how you deploy. 116 of 163 servers run over stdio, meaning the client starts them as a local subprocess on your machine. Only 62 support Streamable HTTP, and 60 publish a hosted remote endpoint you can point at without installing anything.

That imbalance is the single biggest practical limitation of MCP today. Local stdio servers are simple and private, but they cannot be shared across a team, they do not work from a phone or a browser-only client, and every machine needs its own install and its own credentials. The servers with remote endpoints are the ones you can actually centralise:

Server	Transports	Remote endpoint	Package	Weekly downloads
SmartBear MCP	stdio, streamable-http	Yes	npm	14,179
mcp	stdio, streamable-http	Yes	pypi	7,222
Rootly	stdio, sse	Yes	pypi	3,503
Homespun	stdio, streamable-http	Yes	npm	3,208
Rendobar	stdio, streamable-http	Yes	npm	1,044
mcp	stdio, streamable-http	Yes	npm	408
AgentPhone	stdio, streamable-http	Yes	npm	144
mcp	stdio, streamable-http	Yes	pypi	110
BoostedTravel	stdio, streamable-http	Yes	npm, pypi	77
HeyClaude - Claude & AI workflow directory	stdio, streamable-http	Yes	npm	54
World Monitor	streamable-http	Yes	remote only	not reported
mcp-server	streamable-http	Yes	remote only	not reported

Packaging follows the same practical split:

Package registry	Servers
npm	78
pypi	34
mcpb	10
oci	9
nuget	6

Are MCP Servers Secure?

On the evidence here, there is no advisory problem yet: 6 security advisories across all 163 servers, with 108 (66%) showing no known advisory and 50 not checkable because they ship no queryable package.

That is worth stating carefully rather than celebrating. A clean advisory record on a young ecosystem mostly means nobody has looked yet. Compare it with our AI WordPress plugins report, where two in three plugins have a disclosed CVE, not because WordPress plugins are worse but because WordPress has a mature vulnerability-reporting culture and a funded advisory feed. MCP has neither yet.

So the real security question for MCP is not “does this server have a CVE”. It is what the server can do once you give it credentials:

1. Scope the credentials, not the server. Give a database server a read-only role. Give a cloud server a role with the two permissions it needs.
2. Prefer first-party for anything holding production keys. 56% of this list qualifies.
3. Read the tool list before you enable it. A server advertises the actions it can take; that list is the actual permission surface.
4. Test with the MCP Inspector first, before wiring a server into an assistant that will call it autonomously.
5. Treat write access as a deliberate choice. Most useful MCP work is read-only.

Are MCP Servers Actively Maintained?

Yes, more so than any other ecosystem we track. The median server was updated 3 days ago, 142 of 163 (87%) within the last 30 days, 153 within 90, and none have gone more than a year. The oldest is 290 days.

That is what a young ecosystem looks like, and it is a genuine advantage of picking an MCP server today: almost nothing here is abandoned. It also means the picture changes fast, which is why this report carries a pull date rather than pretending to be evergreen.

no servers carry a deprecated flag from the registry. They are still listed, with the flag visible, because a deprecated server does not vanish from the rest of the internet just because a directory hides it.

Best MCP Servers by Category

Nobody needs “the best MCP server”. They need one for Postgres, or Slack, or a browser. Each group below is the top servers in that category by quality score, with weekly downloads, first-party status and update recency alongside so you can overrule the ordering on whichever signal you trust most.

Best Developer Tools MCP servers

Developer-tool MCP servers are the largest and most active category, wiring AI into code, repos, and issue trackers.

These servers connect assistants like Claude and Cursor to GitHub, GitLab, Jira, Linear, and your local toolchain so an AI can read code, open pull requests, and triage issues. It is the most mature MCP category and the one most clients ship support for first.

54 servers in this category, 1,618,259 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Chrome DevTools MCP	1,563,379	First-party	0d ago	93/100
Windows-MCP	18,688	First-party	0d ago	89/100
macOS-MCP	2,975	First-party	1d ago	86/100
CrowdStrike Falcon MCP Server	8,764	First-party	1d ago	85/100
Funplay Unity MCP	not reported	First-party	10d ago	83/100
FunseaAI Unity MCP	not reported	First-party	10d ago	83/100

Best Other MCP servers

Servers across every other niche are joining the MCP ecosystem.

From media and IoT to specialized internal tools, servers of every kind are adopting MCP. This catch-all category captures the breadth of the protocol beyond the obvious developer and data tools.

26 servers in this category, 33,934 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
World Monitor	not reported	First-party	0d ago	94/100
Oh My Posh Validator	not reported	First-party	0d ago	91/100
apify-mcp-server	not reported	First-party	0d ago	90/100
Microsoft Fabric MCP Server	882	First-party	0d ago	89/100
Power BI Modeling MCP Server	14,080	First-party	7d ago	88/100
Jitsu	not reported	First-party	0d ago	85/100

Best Search & Web MCP servers

Search and web MCP servers give an AI fresh, real-world context beyond its training cutoff.

Web search, fetch, and browser-automation servers let assistants look things up, read pages, and drive a browser. They are the backbone of grounded, up-to-date AI answers.

24 servers in this category, 12,794 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Agent Skills Search Server	not reported	First-party	3d ago	93/100
browser-use	library-wide	First-party	1d ago	91/100
exa	not reported	First-party	0d ago	87/100
BoostedTravel	77	First-party	2d ago	86/100
Zotero MCP	7,638	Community	2d ago	84/100
Atlassian Rovo MCP Server	not reported	First-party	11d ago	83/100

Best AI & Memory MCP servers

AI and memory MCP servers give assistants persistent memory and a knowledge base.

Vector stores, knowledge graphs, and note systems let an AI remember context across sessions and ground answers in your own documents. The foundation of RAG-style workflows over MCP.

13 servers in this category, 4,755 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Codebase Memory	4,175	Community	0d ago	96/100
sem	not reported	First-party	20d ago	89/100
Persome	library-wide	First-party	1d ago	83/100
Microsoft NuGet	not reported	First-party	2d ago	82/100
agent-recall	580	Community	3d ago	80/100
ai-context	library-wide	First-party	0d ago	79/100

Best Productivity MCP servers

Productivity MCP servers connect AI to notes, docs, files, and project tools.

Notion, Google Drive, filesystem, calendar, and task-manager servers let an AI read and update your working documents and to-dos. Among the most-installed servers because they touch everyday workflows.

10 servers in this category, 0 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Repowise	library-wide	First-party	0d ago	95/100
mcp-server	not reported	First-party	1d ago	91/100
XcodeBuildMCP	not reported	First-party	2d ago	90/100
Microsoft Learn MCP	not reported	First-party	2d ago	84/100
Svelte MCP	not reported	First-party	0d ago	81/100
docfork-mcp	not reported	First-party	55d ago	79/100

Best Communication MCP servers

Communication MCP servers let AI read and send across chat and email.

Slack, Discord, email, and messaging servers let an assistant summarize threads, draft replies, and post updates. Convenient, but write access to your inbox or channels deserves scrutiny.

8 servers in this category, 294 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Atomic Mail	294	First-party	9d ago	85/100
draw.io	not reported	First-party	4d ago	85/100
emailmd	not reported	First-party	12d ago	83/100
beever-atlas	not reported	First-party	0d ago	80/100
Gmail-MCP-Server	not reported	Community	20d ago	73/100
e2a - email for AI agents	not reported	First-party	0d ago	73/100

Best DevOps & Monitoring MCP servers

DevOps and monitoring MCP servers expose infrastructure and observability data to AI.

Kubernetes, Docker, Terraform, Sentry, Grafana, and incident tools let an assistant inspect deployments, read logs, and surface errors. Powerful for on-call work, where read-only scopes are the safe default.

8 servers in this category, 3,604 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
mockserver	not reported	First-party	0d ago	86/100
Unity-MCP	not reported	Community	4d ago	84/100
Rootly	3,503	First-party	1d ago	82/100
monitor	101	First-party	0d ago	81/100
GoModel	not reported	First-party	0d ago	79/100
Unraid RMCP	library-wide	First-party	2d ago	75/100

Best Databases MCP servers

Database MCP servers let an AI query and inspect your data through a controlled interface.

Postgres, MySQL, SQLite, MongoDB, Redis, and managed platforms like Supabase expose schema and read (sometimes write) access over MCP. Useful for natural-language analytics, but scope the credentials carefully, since you are handing an AI a database connection.

7 servers in this category, 57,911 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
ClickHouse	47,481	First-party	2d ago	88/100
Supabase	not reported	First-party	1d ago	88/100
ArcadeDB MCP Server	not reported	First-party	0d ago	82/100
mcp	7,222	First-party	0d ago	81/100
QueryWeaver	not reported	First-party	1d ago	81/100
Homespun	3,208	First-party	1d ago	79/100

Best Finance MCP servers

Finance MCP servers connect AI to payments, accounting, and market data.

Stripe, accounting, and banking servers let an assistant read transactions, draft invoices, and pull financial data. High-stakes by nature, so keep write access tightly scoped.

5 servers in this category, 0 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Finance Toolkit	library-wide	Community	0d ago	86/100
mcp	not reported	First-party	1d ago	83/100
blockrun-mcp	not reported	First-party	2d ago	81/100
jjlabsio-korea-stock-mcp	not reported	Community	1d ago	73/100
zwldarren-akshare-one-mcp	not reported	Community	146d ago	66/100

Best Cloud MCP servers

Cloud MCP servers expose AWS, Azure, GCP, and edge platforms to AI.

These servers let an assistant query cloud resources, read storage, and manage services. Among the most sensitive in this list, so prefer scoped, read-only credentials and audit every write.

4 servers in this category, 523,063 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Scrapling MCP Server	library-wide	Community	1d ago	92/100
Azure MCP Server	114,702	First-party	0d ago	89/100
mcp	not reported	First-party	3d ago	86/100
Amazon ECS MCP Server	408,361	First-party	179d ago	78/100

Best Design MCP servers

Design MCP servers connect AI to design tools and image generation.

Figma, asset, and image-generation servers let an AI read design files, generate visuals, or produce screenshots. A smaller but fast-growing category as design tools open up MCP access.

4 servers in this category, 1,910 combined weekly downloads.

Server	Weekly downloads	First-party	Updated	Quality
Figma-Context-MCP	not reported	Community	0d ago	88/100
Figma MCP Server	not reported	First-party	1d ago	82/100
ai-test-process-mcp	1,910	Community	0d ago	74/100
mcp	not reported	First-party	59d ago	71/100

The Full List: All 163 MCP Servers Ranked

Every ranked server, ordered by quality score, with the inputs shown so you can check the arithmetic. Click any server to open its repository. Where a value reads “not reported”, the upstream source did not return it and we would rather say so than print a zero.

#	MCP server	Category	Weekly downloads	GitHub stars	First-party	Licence	Updated	Security	Quality
1	Codebase Memory	AI & Memory	4,175	37,981	Community	MIT	0d ago	No known advisory	96
2	Repowise	Productivity	library-wide	4,833	First-party	AGPL-3.0	0d ago	No known advisory	95
3	World Monitor	Other	not reported	79,567	First-party	not reported	0d ago	Not checked	94
4	Chrome DevTools MCP	Developer Tools	1,563,379	48,690	First-party	Apache-2.0	0d ago	Advisory, patched	93
5	Agent Skills Search Server	Search & Web	not reported	23,976	First-party	Apache-2.0	3d ago	Not checked	93
6	Scrapling MCP Server	Cloud	library-wide	72,958	Community	BSD-3-Clause	1d ago	No known advisory	92
7	browser-use	Search & Web	library-wide	108,166	First-party	MIT	1d ago	Advisory, patched	91
8	mcp-server	Productivity	not reported	42,995	First-party	AGPL-3.0	1d ago	Not checked	91
9	Oh My Posh Validator	Other	not reported	23,230	First-party	MIT	0d ago	Not checked	91
10	XcodeBuildMCP	Productivity	not reported	6,202	First-party	MIT	2d ago	No known advisory	90
11	apify-mcp-server	Other	not reported	2,834	First-party	MIT	0d ago	Not checked	90
12	Azure MCP Server	Cloud	114,702	3,546	First-party	MIT	0d ago	Advisory, patched	89
13	Windows-MCP	Developer Tools	18,688	6,664	First-party	MIT	0d ago	Advisory, patched	89
14	Microsoft Fabric MCP Server	Other	882	3,546	First-party	MIT	0d ago	No known advisory	89
15	sem	AI & Memory	not reported	3,279	First-party	Apache-2.0	20d ago	No known advisory	89
16	ClickHouse	Databases	47,481	844	First-party	Apache-2.0	2d ago	No known advisory	88
17	Power BI Modeling MCP Server	Other	14,080	1,050	First-party	MIT	7d ago	No known advisory	88
18	Figma-Context-MCP	Design	not reported	15,604	Community	MIT	0d ago	Advisory, patched	88
19	Supabase	Databases	not reported	2,855	First-party	Apache-2.0	1d ago	No known advisory	88
20	exa	Search & Web	not reported	4,828	First-party	MIT	0d ago	Not checked	87
21	macOS-MCP	Developer Tools	2,975	145	First-party	MIT	1d ago	No known advisory	86
22	BoostedTravel	Search & Web	77	1,643	First-party	not reported	2d ago	No known advisory	86
23	Finance Toolkit	Finance	library-wide	5,199	Community	MIT	0d ago	No known advisory	86
24	mockserver	DevOps & Monitoring	not reported	4,933	First-party	Apache-2.0	0d ago	Not checked	86
25	mcp	Cloud	not reported	4,036	First-party	Apache-2.0	3d ago	Not checked	86
26	CrowdStrike Falcon MCP Server	Developer Tools	8,764	230	First-party	MIT	1d ago	No known advisory	85
27	Atomic Mail	Communication	294	270	First-party	MIT	9d ago	No known advisory	85
28	draw.io	Communication	not reported	5,133	First-party	Apache-2.0	4d ago	Not checked	85
29	Jitsu	Other	not reported	5,010	First-party	MIT	0d ago	Not checked	85
30	voicemode	Other	library-wide	1,312	First-party	MIT	2d ago	No known advisory	85
31	Zotero MCP	Search & Web	7,638	4,565	Community	MIT	2d ago	No known advisory	84
32	strata	Other	not reported	5,786	First-party	Apache-2.0	67d ago	Not checked	84
33	Unity-MCP	DevOps & Monitoring	not reported	3,838	Community	Apache-2.0	4d ago	Not checked	84
34	Microsoft Learn MCP	Productivity	not reported	1,819	First-party	CC-BY-4.0	2d ago	Not checked	84
35	SmartBear MCP	Other	14,179	42	First-party	MIT	0d ago	No known advisory	83
36	Anki MCP Server	Other	606	419	First-party	MIT	16d ago	No known advisory	83
37	mcp	Finance	not reported	1,726	First-party	MIT	1d ago	Not checked	83
38	Persome	AI & Memory	library-wide	1,347	First-party	Apache-2.0	1d ago	No known advisory	83
39	emailmd	Communication	not reported	1,292	First-party	MIT	12d ago	No known advisory	83
40	Atlassian Rovo MCP Server	Search & Web	not reported	945	First-party	Apache-2.0	11d ago	Not checked	83
41	Funplay Unity MCP	Developer Tools	not reported	211	First-party	MIT	10d ago	No known advisory	83
42	FunseaAI Unity MCP	Developer Tools	not reported	211	First-party	MIT	10d ago	No known advisory	83
43	Rootly	DevOps & Monitoring	3,503	45	First-party	Apache-2.0	1d ago	No known advisory	82
44	mcp	Other	not reported	2,597	First-party	Apache-2.0	0d ago	Not checked	82
45	Figma MCP Server	Design	not reported	1,863	First-party	not reported	1d ago	Not checked	82
46	Microsoft NuGet	AI & Memory	not reported	1,553	First-party	not reported	2d ago	No known advisory	82
47	ArcadeDB MCP Server	Databases	not reported	1,065	First-party	Apache-2.0	0d ago	Not checked	82
48	mcp	Databases	7,222	84	First-party	MIT	0d ago	No known advisory	81
49	Auth0 MCP Server	Other	3,143	112	First-party	MIT	0d ago	No known advisory	81
50	monitor	DevOps & Monitoring	101	1,232	First-party	not reported	0d ago	No known advisory	81
51	browserbasehq-mcp-browserbase	Search & Web	not reported	3,416	Community	Apache-2.0	18d ago	Not checked	81
52	claude-real-video	Developer Tools	library-wide	1,973	Community	MIT	4d ago	No known advisory	81
53	QueryWeaver	Databases	not reported	1,060	First-party	AGPL-3.0	1d ago	Not checked	81
54	blockrun-mcp	Finance	not reported	479	First-party	MIT	2d ago	No known advisory	81
55	Svelte MCP	Productivity	not reported	304	First-party	MIT	0d ago	No known advisory	81
56	agent-recall	AI & Memory	580	364	Community	MIT	3d ago	No known advisory	80
57	MCP Fiscal Brasil	Developer Tools	162	137	First-party	MIT	3d ago	No known advisory	80
58	unreal-engine-mcp	Developer Tools	not reported	822	Community	MIT	0d ago	No known advisory	80
59	touchdesigner-mcp-server	Developer Tools	not reported	466	Community	MIT	1d ago	No known advisory	80
60	beever-atlas	Communication	not reported	437	First-party	Apache-2.0	0d ago	Not checked	80
61	postman-mcp-server	Developer Tools	not reported	293	First-party	Apache-2.0	2d ago	No known advisory	80
62	Xquik MCP Server	Other	not reported	176	First-party	MIT	4d ago	Not checked	80
63	adeu	Other	not reported	135	First-party	MIT	0d ago	No known advisory	80
64	Deep Agentic Core MCP	Developer Tools	9,076	not reported	First-party	not reported	2d ago	No known advisory	79
65	OpenTakeoff	Developer Tools	3,260	65	First-party	Apache-2.0	0d ago	No known advisory	79
66	Homespun	Databases	3,208	1	First-party	MIT	1d ago	No known advisory	79
67	brave	Search & Web	not reported	1,358	Community	MIT	2d ago	Not checked	79
68	GoModel	DevOps & Monitoring	not reported	1,043	First-party	MIT	0d ago	Not checked	79
69	docfork-mcp	Productivity	not reported	489	First-party	MIT	55d ago	No known advisory	79
70	gk-cli	Developer Tools	not reported	442	First-party	not reported	3d ago	No known advisory	79
71	ai-context	AI & Memory	library-wide	3	First-party	Apache-2.0	0d ago	No known advisory	79
72	Amazon ECS MCP Server	Cloud	408,361	not reported	First-party	not reported	179d ago	No known advisory	78
73	Unraid MCP	Developer Tools	2,204	121	First-party	MIT	2d ago	No known advisory	78
74	monday.com	Productivity	not reported	417	First-party	MIT	0d ago	Not checked	78
75	spotifyscraper	Developer Tools	library-wide	288	Community	MIT	2d ago	No known advisory	78
76	zoo-mcp	Developer Tools	1,340	8	First-party	MIT	1d ago	No known advisory	77
77	mcp	Developer Tools	408	179	First-party	Apache-2.0	1d ago	No known advisory	77
78	google-surf-mcp	Search & Web	364	276	Community	MIT	2d ago	No known advisory	77
79	Ignite UI MCP Server	Developer Tools	348	136	First-party	MIT	0d ago	No known advisory	77
80	ChiR24-unreal_mcp	Other	not reported	822	Community	MIT	0d ago	Not checked	77
81	ChiR24-unreal_mcp_server	AI & Memory	not reported	822	Community	MIT	0d ago	Not checked	77
82	Memorix	AI & Memory	not reported	613	Community	Apache-2.0	1d ago	No known advisory	77
83	mcp-server	Other	not reported	305	First-party	Apache-2.0	0d ago	Not checked	77
84	Code Pathfinder	Search & Web	library-wide	140	First-party	Apache-2.0	28d ago	No known advisory	77
85	Docmancer	AI & Memory	library-wide	119	First-party	MIT	2d ago	No known advisory	77
86	ClaudeR - RStudio MCP Server	Developer Tools	3,275	314	Community	not reported	4d ago	No known advisory	76
87	Rendobar	Other	1,044	1	First-party	MIT	1d ago	No known advisory	76
88	glade-mcp	Developer Tools	786	187	First-party	MIT	0d ago	No known advisory	76
89	Alfanous - Quranic Search Engine	Search & Web	106	286	First-party	LGPL-3.0	55d ago	No known advisory	76
90	HeyClaude - Claude & AI workflow directory	Search & Web	54	286	Community	MIT	7d ago	No known advisory	76
91	Zapier	Other	not reported	372	First-party	MIT	9d ago	Not checked	76
92	gopeak	Developer Tools	library-wide	239	Community	MIT	25d ago	No known advisory	76
93	mcp-accessibility-scanner	Search & Web	1,726	56	Community	MIT	2d ago	No known advisory	75
94	dash-mcp-server	Search & Web	45	175	Community	MIT	7d ago	No known advisory	75
95	ref-tools-ref-tools-mcp	Other	not reported	1,149	Community	MIT	42d ago	Not checked	75
96	misakanet	Search & Web	library-wide	420	Community	Apache-2.0	0d ago	No known advisory	75
97	office-oxide-mcp	Developer Tools	not reported	155	First-party	GPL-3.0	52d ago	No known advisory	75
98	Funplay Cocos MCP	Developer Tools	not reported	152	First-party	MIT	7d ago	No known advisory	75
99	Unraid RMCP	DevOps & Monitoring	library-wide	121	First-party	MIT	2d ago	No known advisory	75
100	ai-test-process-mcp	Design	1,910	not reported	Community	MIT	0d ago	No known advisory	74
101	Kiwoom Securities MCP Server	Developer Tools	1,663	1	Community	MIT	0d ago	No known advisory	74
102	dex-data	Search & Web	1,581	not reported	First-party	not reported	3d ago	No known advisory	74
103	AgentPhone	Developer Tools	144	112	First-party	MIT	9d ago	No known advisory	74
104	squirrelscan	Other	not reported	252	First-party	MIT	0d ago	Not checked	74
105	build	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
106	cargo	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
107	docker	DevOps & Monitoring	not reported	136	Community	MIT	3d ago	No known advisory	74
108	git	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
109	github	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
110	go	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
111	http	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
112	lint	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
113	make	Productivity	not reported	136	Community	MIT	3d ago	No known advisory	74
114	npm	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
115	pare-build	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
116	pare-cargo	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
117	pare-docker	DevOps & Monitoring	not reported	136	Community	MIT	3d ago	No known advisory	74
118	pare-git	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
119	pare-github	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
120	pare-go	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
121	pare-http	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
122	pare-lint	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
123	pare-make	Productivity	not reported	136	Community	MIT	3d ago	No known advisory	74
124	pare-npm	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
125	pare-python	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
126	pare-search	Search & Web	library-wide	136	Community	MIT	3d ago	No known advisory	74
127	pare-test	Developer Tools	library-wide	136	Community	MIT	3d ago	No known advisory	74
128	python	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
129	search	Search & Web	not reported	136	Community	MIT	3d ago	No known advisory	74
130	test	Developer Tools	not reported	136	Community	MIT	3d ago	No known advisory	74
131	inspeximus	AI & Memory	library-wide	5	Community	MIT	0d ago	No known advisory	74
132	Amicus	Developer Tools	library-wide	not reported	Community	MIT	0d ago	No known advisory	74
133	playwright-stealth	Search & Web	1,203	75	Community	MIT	12d ago	No known advisory	73
134	hustcc-mcp-mermaid	Other	not reported	621	Community	MIT	84d ago	Not checked	73
135	emisar	Other	not reported	420	First-party	not reported	0d ago	Not checked	73
136	SqlServer.Rules	Databases	not reported	237	Community	MIT	3d ago	No known advisory	73
137	Gmail-MCP-Server	Communication	not reported	223	Community	MIT	20d ago	No known advisory	73
138	Glif	Other	not reported	199	First-party	MIT	8d ago	Not checked	73
139	e2a - email for AI agents	Communication	not reported	181	First-party	Apache-2.0	0d ago	Not checked	73
140	Betterlytics	Search & Web	not reported	172	First-party	AGPL-3.0	1d ago	Not checked	73
141	jjlabsio-korea-stock-mcp	Finance	not reported	170	Community	ISC	1d ago	Not checked	73
142	mcp-server	Search & Web	not reported	115	First-party	MIT	8d ago	Not checked	73
143	bettermemory	AI & Memory	library-wide	not reported	Community	MIT	1d ago	No known advisory	73
144	Excalidraw Architect	Developer Tools	555	139	Community	MIT	60d ago	No known advisory	72
145	docfork-mcp	Developer Tools	not reported	489	Community	MIT	55d ago	Not checked	72
146	virustotal	Developer Tools	not reported	142	Community	MIT	75d ago	No known advisory	71
147	mcp	Design	not reported	135	First-party	MIT	59d ago	Not checked	71
148	docs-mcp	Productivity	not reported	114	First-party	not reported	2d ago	Not checked	71
149	fondue-city-mcp	Developer Tools	1,122	not reported	Community	not reported	3d ago	No known advisory	70
150	pinkpixel-dev-web-scout-mcp	Search & Web	not reported	132	Community	Apache-2.0	17d ago	Not checked	70
151	mcp	Developer Tools	110	125	First-party	MIT	148d ago	No known advisory	69
152	DeepSeek MCP Server	Communication	not reported	351	Community	MIT	105d ago	No known advisory	69
153	DottedSign	Developer Tools	not reported	143	First-party	not reported	23d ago	Not checked	69
154	ClawLink	Communication	not reported	117	First-party	AGPL-3.0	0d ago	Not checked	69
155	context-sync	AI & Memory	library-wide	183	Community	MIT	118d ago	No known advisory	67
156	Glean Remote MCP Server	AI & Memory	not reported	164	First-party	MIT	112d ago	Not checked	67
157	Analook - Competitor Intelligence	Developer Tools	not reported	105	Community	not reported	1d ago	Not checked	67
158	oxylabs-oxylabs-mcp	Search & Web	not reported	101	Community	MIT	60d ago	Not checked	67
159	zwldarren-akshare-one-mcp	Finance	not reported	219	Community	MIT	146d ago	Not checked	66
160	svelte-llm-mcp	Other	not reported	160	First-party	MIT	175d ago	Not checked	66
161	shodan	Search & Web	not reported	148	Community	MIT	129d ago	No known advisory	66
162	mcp	Other	not reported	178	First-party	not reported	98d ago	Not checked	64
163	hypertool-mcp	Other	not reported	155	First-party	not reported	290d ago	No known advisory	61

What This Data Says About MCP Right Now

Five conclusions, all checkable against the tables above:

1. MCP is still mostly a local-process protocol. 116 of 163 servers run over stdio, and while 60 now publish a hosted endpoint, that leaves the majority installable only on the machine that runs the AI client. It is shifting, but MCP is still a developer-machine technology more than a team one.
2. First-party is the signal to buy on. 56% of servers are vendor-built and they take 98.9% of all download volume, so the ecosystem is consolidating around official bridges.
3. Adoption is concentrated, but less than published figures suggest. Once library-wide packages are set aside, the most-downloaded MCP server takes 69.3% of attributable volume against a median of 1,663 a week. That is still a steep curve, and it is the honest version of it.
4. The clean security record is a maturity artefact, not a guarantee. 6 advisories across 163 servers, all patched, is not evidence that MCP servers are safe: 50 of them publish nothing an advisory database can be queried against, so their record is unknown rather than clean. Scope every credential you hand a server as if a disclosure were coming.
5. Nothing here is abandoned yet. Median 3 days since update, which is the healthiest maintenance profile of any ecosystem in this series.

“The star counts and the awesome-lists tell you what developers find interesting. What actually decides whether you should install an MCP server is narrower: did the vendor build it, what credentials does it want, and can you run it anywhere other than your own laptop. On that last one the answer is still mostly no, and that is the real state of MCP in 2026, whatever the server count suggests.”
Alston Antony, founder of zplatform.ai and Senior Digital Marketing Manager at Brainstorm Force

Connecting AI to other places instead? The best AI WordPress plugins report carries a full CVE audit, the best Hugging Face models report ranks open-weight models on download data, and the AI Chrome extensions and AI Firefox add-ons reports cover the browser. More sit in best AI tools.

Frequently Asked Questions

What is an MCP server?

An MCP server is a program that exposes tools, data and prompts to AI applications through the Model Context Protocol. It acts as a standardised bridge so an assistant like Claude, Cursor or ChatGPT can read from and act on an external system, a database, a GitHub repo, a Slack workspace, without custom integration code.

What does MCP stand for?

MCP stands for Model Context Protocol, an open standard introduced by Anthropic in November 2024 for connecting AI applications to external tools and data sources.

How does an MCP server work?

An MCP host (the AI app) runs a client that connects to one or more servers over JSON-RPC. Servers expose three primitives: tools the AI can call, resources it can read, and prompts it can reuse. The two transports are stdio for local servers and Streamable HTTP for remote ones. In this list 116 of 163 servers use stdio.

What are the best MCP servers?

By weekly package downloads the leader is Chrome DevTools MCP at 1,563,379. By our composite quality score, which weighs maintenance, trust and security alongside adoption, the leader is Codebase Memory at 96/100. Both full rankings are above, and the category sections narrow it to the job you actually have.

Where can I find a list of MCP servers?

This page is one, built from the official MCP Registry and enriched with adoption, maintenance and security data. The canonical source is the registry itself at registry.modelcontextprotocol.io; this report adds the comparison layer on top so you can rank and filter rather than only browse.

Are MCP servers free?

Most are free and open-source, and you install them at no cost. Some connect to paid third-party services (a database host, a SaaS API) that need your own account and API key, but the server software itself is typically free. Every row shows the install method and package registry.

What are the most popular MCP servers?

By weekly downloads: Chrome DevTools MCP (1,563,379), Amazon ECS MCP Server (408,361) and Azure MCP Server (114,702). Popularity clusters around Developer Tools, which takes 71.7% of all download volume from 54 servers.

What is the difference between an MCP server and an API?

An API is a service-specific interface, so every integration is bespoke. An MCP server wraps a system, which may itself sit on an API, in the standard Model Context Protocol, so any MCP-capable client can use it the same way. MCP is the universal adapter; an API is the underlying outlet.

MCP versus RAG: what is the difference?

RAG retrieves relevant text and injects it into a prompt. MCP is a connection protocol that lets an AI call live tools and read data sources directly, including taking actions. They are complementary: an MCP server can power retrieval for RAG, but MCP also covers actions RAG alone does not.

What are the best MCP servers for Claude Code?

Every server in this list works with Claude Code and ships a copy-paste “claude mcp add” command. For coding specifically the useful groups are developer tools, filesystem, databases, and browser or search servers. Use the Developer Tools section above, which is the largest category at 54 servers.

Which AI tools support MCP servers?

Claude Desktop, Claude Code, Cursor and VS Code all support MCP, and every server in this build ships install snippets for those four. ChatGPT, Gemini and Windsurf have also added MCP support. The client table above shows snippet coverage across all 163 servers.

What is the official MCP Registry?

The official MCP Registry (registry.modelcontextprotocol.io) is the canonical index of MCP servers, where publishers register under an ownership-verified reverse-DNS namespace. That verification is what makes the first-party badge in this report meaningful, and this list is built from it.

How do I install an MCP server in Claude Desktop?

Open Claude Desktop settings, edit claude_desktop_config.json, add the server under “mcpServers” with its command, args and any required environment variables, then restart Claude Desktop. Every row in the full list above provides the exact JSON, along with the equivalent for Cursor, VS Code and Claude Code.

How many MCP servers are there?

The official registry lists thousands and grows daily. This report ranks the 163 that clear the quality gate and adoption floor: a description, a working install method, and real adoption, as of 7 August 2026.

Who created MCP?

Anthropic created the Model Context Protocol and released it as an open standard in November 2024. It has since been adopted by many AI vendors, with an open-source community maintaining the specification and the official registry.

Methodology and How to Cite This Data

Sources: server identity, install methods and status from the official MCP Registry; stars, licence and last-commit recency from the GitHub REST and GraphQL APIs; weekly downloads from npm and PyPI; security advisories from the GitHub Advisory Database. 6 snapshots retained for trend history.

Sample: 163 servers, 2,256,524 combined weekly downloads, 11 categories, 92 first-party, 163 with install snippets across 4 AI clients.

Ranking formula: Quality Score = 35% adoption (GitHub stars and weekly package downloads, log-scaled) + 25% maintenance (update recency and release cadence) + 15% growth (30-day trend from our own snapshots) + 15% trust (first-party namespace, open-source licence, public repository) + 10% security (known unpatched advisories).

Inclusion: automatic. Every registry server with a description, a working install method and real adoption. Deleted servers excluded, deprecated servers kept and badged. No hand-picking, no paid placement.

Download attribution rule: a server’s download count is only counted when the npm or PyPI package it points at is the MCP server itself, which we test by requiring the package name to reference MCP. Registry entries pointing at a wider general-purpose library are excluded from every download total and marked library-wide in the full list, because that count measures the library rather than the server. In this build that excluded 26 servers and 13,655,940 weekly downloads, 86% of the raw sum. The rule is deliberately mechanical, so it is reproducible and it errs toward excluding a genuine MCP package rather than inflating a total. It cannot split a single shared package across a family of servers.

Stated limitations: adoption is a proxy. MCP has no universal usage metric, so package downloads and stars stand in for it and are never presented as install counts. Where the upstream GitHub fetch does not return data for a server, its stars, licence and release history read “not reported” rather than zero, and any trend figure that would imply a total collapse as a result is suppressed rather than published. Coverage in this build: 156 of 163 servers with a star count, 146 with a licence, 44 with attributable download data, 120 with a usable trend (0 suppressed as artefacts). Security is “not checked” for servers with no queryable package rather than reported as clean.

Cite as: zplatform.ai, “Best MCP Servers: The Complete Ranked List,” data pulled 7 August 2026.

Every number here traces to a row in the tables above, and where a number is missing this page says so instead of guessing. That is the whole point of publishing the method next to the ranking: the best MCP server for you depends on what you are connecting and what credentials you are willing to hand over, and you should be able to check the working rather than take my word for it.