Best MCP servers for Claude Code in 2026

TL;DR. The best MCP servers for Claude Code in 2026 are the official, remote ones that do one job well: GitHub for repos and CI, Supabase or Hatchable for a database and a live app, Playwright for a browser, Context7 for current docs, Sentry for errors, Linear and Notion for the team's work, Stripe for payments, Cloudflare for edge infrastructure. Add a remote server with claude mcp add --transport http <name> <url>, sign in with /mcp, and keep the list short.

How this list of Claude Code MCP servers was chosen

Claude Code can connect to hundreds of MCP servers, and most "best MCP servers for Claude Code" lists are forty links long. This one is ten, picked on four rules. The vendor's own server beats a community fork, because it tracks the product's API. Remote Streamable HTTP with OAuth beats a local process that needs a token in a file, because there is nothing to install and nothing to leak. Each server has to do one job you can name. And each has a documented Claude Code install, so the one-liner below is the vendor's, not ours.

General characterisations based on public information as of August 2026; check each vendor for current details.

First time? How to add an MCP server to Claude Code covers the command, scopes and troubleshooting, and what MCP is covers the protocol.

The best MCP servers for Claude Code in 2026

1. GitHub MCP server (source control and CI)

Good for: issues, pull requests, code search, and Actions logs pulled into the session so Claude Code can fix a red build without you pasting anything. GitHub's official server, hosted by GitHub, with toolsets you can narrow and a read-only mode.

claude mcp add --transport http github https://api.githubcopilot.com/mcp/ \
  --header "Authorization: Bearer YOUR_GITHUB_PAT"

Caveat: as of August 2026, GitHub's own install guides for Claude Code and Cursor use a personal access token rather than the OAuth flow that VS Code gets. Make it a fine-grained token scoped to the repos you need, and turn off the write toolsets you will not use.

2. Supabase MCP server (hosted Postgres)

Good for: an app whose backend already lives on Supabase. Claude Code can inspect the schema, run SQL, write migrations, manage branches, deploy edge functions and read logs through one remote endpoint with OAuth sign-in.

claude mcp add --transport http supabase "https://mcp.supabase.com/mcp?read_only=true"

Caveat: it runs real queries against a real project. Supabase's docs recommend pointing it at a development project rather than production and keeping read_only=true on until you need writes. Treat every write tool as a deploy.

3. Hatchable (build, deploy, host, and a database)

Good for: the case where Claude Code should build the whole app and put it live, not just talk to a backend you already run. One OAuth sign-in at https://hatchable.com/mcp, no API key, and Claude Code can create a project, write the files, deploy, and hand back a live URL at your-app.hatchable.site. Every project comes with its own private Postgres database, sign-in for the app's users, email, scheduled jobs and file storage. The free plan is unlimited private projects and one published app, no card.

claude mcp add --transport http hatchable https://hatchable.com/mcp

Caveat: it hosts web apps written in HTML, JavaScript, SQL and TOML on its own platform. It is not a Python or Docker host and does not run long-lived processes, so a Flask service or a Discord bot belongs elsewhere. The full tool list is on the MCP feature page, and the step-by-step is in connect Claude Code to Hatchable.

4. Playwright MCP (browser automation)

Good for: letting Claude Code check its own UI work. It navigates, clicks, fills forms and takes screenshots through accessibility snapshots, so the model reads the page structurally instead of guessing from pixels. Microsoft maintains it.

claude mcp add playwright -- npx @playwright/mcp@latest

Caveat: this is a local stdio server. It needs Node.js 20 or newer, launches a real browser on your machine (headed by default, --headless to hide it), and keeps a persistent profile unless you pass --isolated. A browser that remembers your logins is one the agent can use; mind where you point it.

5. Context7 (current library docs)

Good for: version-specific documentation pulled into context at the moment Claude Code needs it, which tends to cut the "that API changed two releases ago" class of bug. Remote endpoint, optional API key.

claude mcp add --transport http context7 https://mcp.context7.com/mcp

Caveat: coverage depends on whether a library is indexed, and Context7's docs describe tighter rate limits without a key. A second opinion, not a substitute for reading the source when it matters.

6. Sentry (errors and performance)

Good for: closing the loop on production bugs. Claude Code pulls the latest unhandled exception, reads the stack trace and release, and proposes the fix in the same session. Sentry's hosted server uses OAuth and can be scoped to one org or project in the URL.

claude mcp add --transport http sentry https://mcp.sentry.dev/mcp

Caveat: only useful for apps that already report to Sentry, and error payloads often carry user data that then flows into the model's context. Scope the URL to the project you are working on.

7. Linear (issue tracking)

Good for: "implement ENG-4521" without opening a browser. Claude Code reads the issue, updates status, leaves comments and creates follow-ups; Linear expanded the hosted server to initiatives, milestones and project updates in early 2026.

claude mcp add --transport http linear https://mcp.linear.app/mcp

Caveat: writes land in your real workspace. Linear documents a read-only endpoint at /mcp/readonly for sessions that only need to read, and its older SSE endpoint is deprecated in favour of /mcp.

8. Notion (specs and internal docs)

Good for: searching and reading the PRD, the runbook or the meeting notes that explain why the code is the way it is, and writing pages back. Notion's hosted server is OAuth only.

claude mcp add --transport http notion https://mcp.notion.com/mcp

Caveat: Notion's docs say the OAuth flow must be completed interactively, and the server can read whatever the signed-in account can read. Sign in with the account whose visibility you intend to grant.

9. Stripe (payments)

Good for: building a payments integration with the real API in reach: search Stripe's docs, read customers, subscriptions and invoices, create products, prices and payment links. OAuth by default, or a restricted key via a header.

claude mcp add --transport http stripe https://mcp.stripe.com

Caveat: write tools move real money (refunds, subscription changes). Stripe's own docs tell you to keep human confirmation of tools on and to be cautious running it alongside other servers, because a prompt injection elsewhere can reach it. Work in a sandbox first.

10. Cloudflare (edge infrastructure and docs)

Good for: Workers and Pages projects. Cloudflare ships a set of small remote servers: documentation for accurate answers about its platform, Workers Bindings to create KV, D1 and R2 resources, Observability for logs and analytics.

claude mcp add --transport http cloudflare-docs https://docs.mcp.cloudflare.com/mcp

Caveat: the docs server needs no sign-in; the account-level servers (bindings, observability and the rest) use OAuth and each adds its own tools. Add the two or three you use, not the full catalogue.

Claude Code MCP servers compared

How to pick MCP servers for Claude Code

Start from a gap, not a list. The right server is the one whose data you keep pasting into the chat. If that is stack traces, add Sentry. If it is "what does the ticket say", add Linear. If Claude Code builds things that then sit on your laptop, add a server that can deploy.

Prefer remote and OAuth. A remote HTTP server has nothing to install and nothing to keep running, and OAuth keeps credentials out of config files you might commit. Use /mcp inside a session (or claude mcp login <name> from your shell) to complete sign-in, and claude mcp list to confirm the status is connected.

Pick the scope on purpose. The default is local to the current project. --scope user makes a server available everywhere; --scope project writes it to .mcp.json so the team shares it, and Claude Code asks for approval before using a project file.

Keep the tool budget honest. Every server adds tool descriptions to the context and every credential widens what an agent can touch. Four to six servers is plenty for most work; Claude Code's tool search softens the context cost but not the blast radius of a mis-aimed write.

Two sensible starter kits. Maintaining an existing codebase: GitHub, Sentry, Linear, Context7, Playwright. Shipping new apps from a prompt: Hatchable for build, deploy and database, GitHub for the repo, Playwright to check the result. The same servers work in Cursor and Codex; only the install command changes.