Best MCP servers in 2026

A reference MCP server that lets assistants read, write, and manage approved local directories with explicit boundaries and safe path handling. It is the canonical example for any agent that needs structured access to a developer machine.

The official GitHub MCP server exposes repositories, issues, pull requests, code search, and Actions context to coding agents. It is published and maintained by GitHub and is the most widely adopted MCP server in production.

Connect AI agents to PostgreSQL with schema inspection, read-only query controls, and concise table context for safer database exploration. The reference implementation also supports restricted write access for migrations.

A browser automation MCP server built on Playwright. Agents can navigate pages, click elements, fill forms, capture screenshots, and inspect the DOM with deterministic waits and accessibility-aware locators.

Expose Brave Search as tools that any MCP-compatible client can call. Returns web, news, and local results with safe-search controls and rich snippets that agents can summarize without scraping.

Search channels, summarize threads, and let agents prepare team updates using governed access to Slack workspaces. Supports read-only and scoped-write modes for safe agent behavior.

Give agents access to Google Maps places, directions, geocoding, and elevation data. Useful for travel planning, real estate, logistics, and any workflow that needs structured location context.

A knowledge-graph based memory layer that lets agents remember entities, relationships, and observations across sessions. The canonical implementation uses a local JSON store for zero-config persistence.

A Puppeteer-based MCP server for scraping, screenshotting, and interacting with web pages. A practical alternative to Playwright when teams already standardize on the Puppeteer ecosystem.

Browse, search, and edit Notion pages and databases through MCP. Ideal for agents that need to read product specs, update docs, or summarize team wikis.

A simple HTTP fetch MCP server that returns cleaned-up HTML, Markdown, or JSON. It is the recommended tool for agents that need to read a page without a full browser stack.

Expose common Git operations to MCP clients: status, log, diff, blame, branch management, and commit history search. A perfect companion to coding agents that work on local clones.

Surface Sentry issues, releases, and stack traces inside agent sessions so debugging tools can propose fixes against real error context. Read-only by default with optional write scopes.

Expose core AWS services to MCP clients: S3 buckets, Lambda functions, CloudWatch logs, and IAM roles. Designed with strict permission scopes for safe agent operations.

Connect Linear to MCP clients so agents can read sprint state, summarize project health, and draft issue updates. Supports team-scoped tokens for safe per-workspace access.

Expose Stripe customers, invoices, subscriptions, and payment intents to MCP clients with strict, auditable scopes. Perfect for support agents that need billing context without opening the dashboard.

The official Supabase MCP server exposes database, auth, storage, and edge function tools. Read-only by default with optional write access for migrations and schema introspection.

Expose Chrome DevTools Protocol commands through MCP. Agents can capture console output, profile performance, manipulate cookies, and execute JavaScript against a live Chrome instance.

Connect Figma files, frames, components, and design tokens to MCP clients. Useful for agents that generate UI from design system primitives or summarize visual specifications.

Expose Cloudflare platform primitives to MCP clients: Workers deployments, KV namespaces, R2 buckets, and D1 databases. A foundation for agents that deploy and operate cloud infrastructure.

Query local SQLite databases through MCP. A lightweight option for desktop tools, browser extensions, and any agent that needs a zero-setup SQL backend.

Bridge JetBrains IDEs to MCP clients with project awareness, refactoring actions, and code intelligence. The companion plugin works with IntelliJ, PyCharm, WebStorm, and GoLand.

Query MongoDB collections, run aggregations, and inspect schema through MCP. A practical fit for content platforms, catalogs, and any document-shaped data product.

Browse and update Jira issues, sprints, and boards from MCP clients. Built for engineering teams that plan, track, and ship work in Jira Cloud.

Read channels, post messages, and manage roles through Discord from MCP clients. A natural fit for community-driven agents and developer operations workflows.

Generate images through EverArt from MCP clients. Useful for design agents that need to produce visual assets inline with a conversation or workflow.

Send transactional emails through Resend from any MCP-compatible client. Useful for agents that need to notify users, send reports, or trigger follow-ups from a workflow.

Read and write Redis keys, manage TTLs, and publish events from MCP clients. Useful for agents that need ephemeral state, rate limiting, or queue coordination.

Search, read, and edit Confluence pages and spaces from MCP clients. A practical fit for internal documentation, onboarding guides, and engineering handbooks.

Send SMS, manage phone numbers, and place voice calls through Twilio from MCP clients. A practical tool for agents that coordinate notifications, alerts, or two-factor flows.

Manage Docker containers, images, and Compose stacks from MCP clients. A practical primitive for agents that provision services and run side-effect tools.

Read and update Salesforce records, opportunities, and accounts from MCP clients. A foundation for enterprise agents that operate on a CRM-first data model.

Manage Vercel projects, deployments, and environment variables from MCP clients. Useful for agents that ship frontend code and need to coordinate deploys.

Read and update Airtable bases, tables, and views through MCP. A natural fit for ops teams that keep customer, content, or campaign data in Airtable.

Surface Datadog metrics, logs, and APM traces inside MCP clients. A foundation for SRE and observability agents that reason about production systems.

Expose HubSpot CRM objects to MCP clients: contacts, companies, deals, and tickets. Useful for sales and support agents that need real customer context.

Read and edit Obsidian vaults through MCP. A natural fit for personal knowledge management agents and research workflows built on local Markdown notes.

Read and operate Kubernetes resources from MCP clients: pods, deployments, services, and logs. A foundation for platform agents that manage cloud-native workloads.

Read and resolve secrets from 1Password vaults through MCP. Designed for agents that need scoped access to credentials, API keys, and infrastructure secrets.

Create, update, and complete Todoist tasks from MCP clients. A natural fit for personal productivity agents and team coordination workflows.