Best MCP Servers for Claude Code in 2026 (Ranked and Tested)

The best MCP servers for Claude Code in 2026 connect Claude to systems it cannot reach on its own: GitHub, your database, a real browser, live library docs, and the web. A short, well-chosen set works better than a long one, because every server adds tools to the agent’s budget and every credential widens the surface you have to trust. This guide ranks the servers worth installing, gives the exact claude mcp add command for each, and flags the popular picks that Claude Code already makes redundant.

Best MCP servers for Claude Code: ranked comparison

MCP server	Best for	Install command	Context cost	Notes
GitHub	Issue and PR workflow	claude mcp add --transport http github https://api.githubcopilot.com/mcp	High	Official hosted server, the largest toolset here
Context7	Up-to-date library docs	claude mcp add context7 -- npx -y @upstash/context7-mcp	Low	Two tools, cuts API hallucination
Playwright	Browser automation and E2E	claude mcp add playwright -- npx @playwright/mcp@latest	Medium	Microsoft official, drives the accessibility tree
Postgres	Database schema and queries	claude mcp add postgres -- npx -y @modelcontextprotocol/server-postgres <read-only-dsn>	Low	Use a read-only role
Exa	Live web search and research	claude mcp add exa -- npx -y exa-mcp-server --api-key KEY	Low	Semantic search built for agents
Sentry	Production error triage	claude mcp add --transport http sentry https://mcp.sentry.dev/mcp	Medium	Official hosted server, auth in the browser
Linear	Tickets and project status	claude mcp add --transport sse linear https://mcp.linear.app/sse	Medium	Official remote server
Slack	Team comms and async updates	claude mcp add slack -e SLACK_BOT_TOKEN=xoxb-... -- npx -y @modelcontextprotocol/server-slack	Medium	Good for long-run notifications
Sequential Thinking	Structured step planning	claude mcp add seq -- npx -y @modelcontextprotocol/server-sequential-thinking	Low	Situational, Claude already reasons in steps
Filesystem	Extra directory access	claude mcp add fs -- npx -y @modelcontextprotocol/server-filesystem <dir>	Low	Usually redundant with built-in file tools

If you use Nimbalyst, the built-in tracker MCP server adds a low-cost planning surface on top of any of these. The full list of useful Claude Code MCP servers runs to dozens, but the ten above cover most working developers, and the first five cover most of the value.

The “essential 5” everyone recommends, tested against Claude Code

Most “best MCP server” lists converge on the same five: Filesystem, GitHub, Context7, Playwright, and Sequential Thinking. Three of them earn a place in a Claude Code setup. Two are mostly redundant with what Claude Code already ships, and installing them anyway just crowds the tool list.

• GitHub. Keep it. The highest-impact server for most developers, covered in detail below.
• Context7. Keep it. Two tools, low cost, and it noticeably reduces wrong-API-signature errors on unfamiliar libraries.
• Playwright. Keep it for frontend work. A real browser the agent can drive to verify its own UI changes.
• Filesystem. Skip it for most Claude Code users. Claude Code already has built-in file tools (Read, Edit, Write, Glob, Grep) plus permission rules and sandboxing. If your goal is keeping Claude away from secrets, use permissions.deny, sandboxing, or a worktree. Install the Filesystem server only when you specifically want to expose additional directories through MCP or share one server config across several MCP clients.
• Sequential Thinking. Situational. It adds a single tool that gives the agent a scratchpad for step-by-step reasoning. Claude models already plan in steps and support extended thinking, so the lift on Claude Code is smaller than on models without native planning. Try it on gnarly multi-step tasks and drop it if you do not see better results.

The same logic rules out the Memory MCP server for a default setup. Claude Code already has two built-in memory systems: CLAUDE.md for persistent instructions and auto memory for learned project notes. A separate memory server earns its place only when you want a shared memory layer across different agents or clients.

The context cost and security tradeoff of running many servers

Two hidden costs grow with every server you add.

Context and tool budget. Each server’s tools are described to the model and count against its tool budget. A long tool list makes the agent slower to choose and more likely to reach for the wrong tool. The cost is not uniform: Context7 adds two tools, while the GitHub server exposes dozens across issues, PRs, Actions, and code scanning. Prefer servers with a tight, relevant toolset, and use read-only modes and toolset filters (the GitHub server supports both) to trim what the agent sees.

Credentials and blast radius. Every server is a token and a permission surface. A write-capable Postgres DSN or an admin-scoped GitHub token is a footgun the moment the agent misfires. Three habits keep this in check:

1. Read-only roles wherever the work allows it (databases especially).
2. Fine-grained tokens scoped to the exact repos, projects, or channels in play.
3. User-scope config for secrets, so tokens live in your home directory and never land in a repo.

A practical rule: three to six servers for most developers. Add one per project when a real need shows up, and remove any server the agent never calls.

Best MCP server by use case

If you want the pick for one specific job rather than a full stack:

Use case	Pick	Why
GitHub workflow	GitHub (official hosted server)	Issues, PRs, code search, and Actions in one server
Database work	Postgres with a read-only DSN	Schema reading and safe query analysis
Browser automation and E2E	Playwright (Microsoft)	Real browser, accessibility-tree driven, self-verifies UI
Web search and research	Exa (or Brave for free)	Semantic retrieval tuned for agents
Library and API docs	Context7	Version-aware docs, fewer hallucinated signatures
Tickets and planning	Linear (or Nimbalyst Tracker in-app)	Reads and updates project state
Production errors	Sentry	Stack traces and issue triage in context

How to evaluate an MCP server

Before installing anything, three questions:

1. Does it expose tools the agent will actually use? Tools the agent ignores still count toward its tool budget and slow things down. Five well-chosen servers work better than twenty.
2. What credentials does it need, and at what scope? A server that needs admin tokens in your production database is a server you have to operate carefully. Use read-only roles wherever possible.
3. Is it actively maintained? The MCP standard is young, and a chunk of the early reference servers have since moved to their vendors or been archived. Prefer servers with commits in the last three months, and prefer a vendor’s official server over a stale community fork.

The detailed list

1. GitHub MCP server

The single highest-impact install for most developers. GitHub now ships an official MCP server, and the cleanest path is the hosted remote endpoint:

claude mcp add --transport http github https://api.githubcopilot.com/mcp

Authenticate with OAuth on first use, or pass a fine-grained personal access token as a header with -H "Authorization: Bearer ghp_...". The older @modelcontextprotocol/server-github npm package is deprecated in favor of this official server. Scope the token to the repos and permissions Claude Code actually needs, and for team projects use a service account so the agent’s actions are auditable separately from your own. The server supports read-only mode and toolset selection, which is the easiest way to keep its large tool list from crowding the agent.

What it unlocks: “Find the issues tagged blocking. Write a PR that closes the top one. Open it as a draft, link the issue, and request review from the owner of the affected module.”

2. Context7 MCP server

Context7 keeps live, version-aware documentation for hundreds of libraries. The agent fetches the right docs for the right version on demand instead of guessing.

claude mcp add context7 -- npx -y @upstash/context7-mcp

Two tools, low context cost, broadly useful. Context7 also offers a hosted remote endpoint if you would rather not run it locally.

What it unlocks: Less hallucination on library APIs. Instead of guessing the current React or Prisma signature, the agent looks it up.

3. Playwright MCP server

Microsoft maintains the official Playwright MCP server. It drives a real browser through the accessibility tree rather than screenshots, which keeps its context footprint lower than image-based automation.

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

ExecuteAutomation ships a popular community alternative, @executeautomation/playwright-mcp-server, if you prefer screenshot-based flows.

What it unlocks: The agent can verify its own UI changes by actually using the app. “Add the new payment form. Then open the checkout page in the browser, fill it out, and confirm the success state appears.”

4. Postgres MCP server

For database-backed projects, exposing the database to Claude Code transforms what it can do: schema reading, query writing, data inspection, and (if you choose) mutations.

claude mcp add postgres -- npx -y @modelcontextprotocol/server-postgres postgresql://readonly@localhost/mydb

Strong recommendation: use a read-only role. Postgres MCP with a read-only DSN is safe and fast. Postgres MCP with a write-capable DSN is a footgun unless you trust the agent fully and have backups. The ecosystem also has actively maintained community options such as Postgres MCP Pro if you want query analysis and stricter write controls.

What it unlocks: “Read the schema. Find the slowest query in the slow_log table. Write a candidate index. Run EXPLAIN ANALYZE before and after.”

5. Exa MCP server (web search)

For live web research without leaving the agent, Exa is the most-used search server for coding agents in 2026. Its API is built around semantic queries and structured results rather than keyword matching.

claude mcp add exa -- npx -y exa-mcp-server --api-key YOUR_EXA_API_KEY

Pick one search server and stop. Running Exa plus Brave plus Tavily together is a common reason the agent picks the wrong tool. Brave Search is a solid free, privacy-first alternative with an independent index, and Tavily is strong for research-style multi-step queries. Claude Code also has a built-in WebSearch tool, so add a search server mainly when you want a specific index or richer retrieval.

What it unlocks: “Pull the three most recent posts on this library’s breaking change, summarize the migration steps, and check them against our current version.”

6. Sentry MCP server

If you ship to production, Sentry wires error context into Claude Code. It reads recent issues, groups by frequency, and pulls stack traces.

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

Sentry runs an official hosted MCP server, so you authenticate through Sentry rather than managing a token in a config file.

What it unlocks: “Look at this morning’s top three Sentry errors. Find the underlying bug in the code. Open a fix PR for the worst one.”

7. Linear MCP server

For teams that live in Linear, this turns Claude Code into a participant in the planning system. It reads tickets, updates status, leaves comments, and creates new issues.

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

Linear ships an official hosted MCP server, so you authenticate in the browser on first connect.

What it unlocks: “Read all open bugs assigned to me. Pick the one with the smallest scope. Open a fix branch, write the patch, and link the Linear ticket from the PR.”

8. Slack MCP server

For teams that coordinate in Slack. Claude Code can read channel history, post messages, and reply in threads.

claude mcp add slack -e SLACK_BOT_TOKEN=xoxb-... -e SLACK_TEAM_ID=T0123... -- npx -y @modelcontextprotocol/server-slack

What it unlocks: Async status updates. The agent can post when a long-running task finishes, and you can reply with follow-up instructions without leaving Slack.

9. Sequential Thinking MCP server

A single tool that gives the agent a structured scratchpad for step-by-step reasoning.

claude mcp add seq -- npx -y @modelcontextprotocol/server-sequential-thinking

Situational for Claude Code, since Claude models already plan in steps and support extended thinking. Worth trying on hard multi-step tasks, and worth removing if it does not change the output.

What it unlocks: A more explicit plan-then-act loop on tangled problems, at the cost of one extra tool.

10. Nimbalyst Tracker MCP server

If you use Nimbalyst, the built-in tracker MCP server gives Claude Code direct access to your in-app plans, tasks, and bugs. The agent can create, update, and close tracker items as part of normal work, with the changes appearing in real time on your kanban board.

What it unlocks: Plans, tasks, and bugs that stay in sync without the agent ever leaving the workspace.

Honorable mentions

• Brave Search for free, privacy-first web research with an independent index.
• Notion for teams whose source-of-truth lives there instead of Linear.
• Stripe for SaaS work that touches billing.
• Cloudflare for edge and DNS workflows.
• Figma for design-to-code handoffs.

The MCP ecosystem doubles every few months. Bookmark the official MCP servers repository for the canonical list.

A sensible starter pack

If you want a day-one setup to install and tune later:

claude mcp add --transport http github https://api.githubcopilot.com/mcp
claude mcp add context7 -- npx -y @upstash/context7-mcp
claude mcp add --transport sse linear https://mcp.linear.app/sse

Add Slack if your team lives there. Add Postgres, Sentry, Playwright, or a search server when the project demands them. Resist installing servers you do not have a clear use for. Each one expands the agent’s tool list, and a bloated tool list hurts the agent’s decision quality.

Claude Code MCP servers in Nimbalyst

Nimbalyst is the open-source visual workspace for building with Codex, Claude Code, and more. The most useful extra MCP server to pair with Nimbalyst is usually GitHub, because Nimbalyst already gives you the visual session layer, tracker, and review workflow. The next-best adds are whatever connect Claude to systems outside the workspace: Linear for tickets, Postgres or Sentry for backend work, and Playwright for UI verification. The MCP configuration you keep in ~/.claude/settings.json still carries over into Nimbalyst sessions, so once you wire those up they show up in both places. For the full setup walkthrough, see the Claude Code MCP setup guide.

Frequently Asked Questions

How many MCP servers should I run?

Three to six is the sweet spot for most developers. Start with GitHub plus one or two project-specific servers, then add Context7 if you frequently work across unfamiliar stacks. Past ten servers, the agent’s tool list gets crowded and its tool choice degrades without much added benefit.

Do MCP servers slow Claude Code down?

They can. Every server adds its tools to the agent’s context and tool budget, which gives the model more options to weigh on each step and can slow tool selection. Low-tool servers like Context7 cost little, while large ones like the GitHub server add a lot. Keep the set small, use read-only and toolset filters where a server offers them, and remove servers the agent never calls.

What is the best MCP server for GitHub?

The official GitHub MCP server. Add it through the hosted endpoint with claude mcp add --transport http github https://api.githubcopilot.com/mcp, authenticate with OAuth or a fine-grained token, and Claude Code can work issues, pull requests, code search, and Actions. Scope the token narrowly and use a service account for team repos so the agent’s actions stay auditable.

What is the best MCP server for Claude Code overall?

GitHub, for most developers. It turns Claude Code from a code generator into a participant in the issue and pull request workflow. If you use Nimbalyst, GitHub is also the cleanest complement, since Nimbalyst already covers the visual session and review layer.

Do I need the Filesystem MCP server for Claude Code?

Usually no. Claude Code already ships built-in file tools (Read, Edit, Write, Glob, Grep) with permission rules and sandboxing. Install the Filesystem server only when you specifically want to expose extra directories through MCP or reuse one server config across several MCP clients.

Related Reading

• Claude Code MCP Setup Guide
• Best MCP Clients in 2026
• Claude Code Subagents Guide
• Claude Code Skills Guide
• Best AI IDEs for Claude Code. Where to host and read Claude Code’s work.
• Best Claude Code GUI in 2026. Five Claude Code GUI tools compared.
• Best AI diagram tools for engineers and Claude Code workflows, including the Excalidraw MCP integration.