Guides6 min read

Best MCP Servers for Zed in 2026: Supercharge the AI-Native Editor

Zed is the AI-native editor built for speed. MCP servers extend your AI assistant beyond the editor — giving it access to your database, repositories, web search, and terminal context. Here's what to add.

By MyMCPTools Team·

Zed is built from the ground up for AI-assisted development — collaborative, fast, and designed to integrate language models directly into the editing experience. With Zed's native AI assistant panel and extension ecosystem, adding MCP servers to your workflow is the natural next step: they give the AI behind Zed the real-world context it needs to reason precisely about your project.

This guide covers the best MCP servers to pair with Zed — whether you're using Claude, GPT-4, or another provider through Zed's AI panel — to build a development environment where your AI assistant knows your codebase, your repositories, and your tools as well as you do.

How MCP Fits Into the Zed Workflow

Zed's AI assistant is powerful out of the box for code completion and inline editing — but MCP servers expand what the AI can know and do. Configure MCP servers in your Claude Desktop or any MCP-compatible client running alongside Zed, and your AI assistant gains structured access to file systems, databases, GitHub, and web search. The result: AI that can act on real project state, not just respond to what you paste into the chat.

1. Filesystem MCP Server — Your Project, AI-Accessible

The filesystem MCP server is the foundation. While Zed's editor already surfaces open files to the AI context window, the filesystem MCP server gives your AI assistant the ability to navigate your entire project — reading any file, searching directory structures, and writing changes — outside the constraints of what's currently open in Zed.

Key capabilities:

  • Read any file in your project directory without manually opening it
  • Search file contents across the full codebase
  • Write and update files as part of multi-step AI workflows
  • Respect configurable access boundaries to protect sensitive directories

Best for: Every Zed developer. The filesystem server extends Zed's AI context beyond the current buffer to the full project scope.

2. Git MCP Server — Commit History and Diff Context

Zed has built-in Git integration for staging and committing, but the Git MCP server gives your AI assistant access to the historical context that makes a difference: commit messages, blame output, branch history, and diff summaries. Your AI can explain why code was written a certain way — not just what it does.

Key capabilities:

  • Read commit history, messages, and authorship
  • Access diff output between branches or commits
  • Query blame data to understand code provenance
  • Navigate branches and tags programmatically

Best for: Debugging sessions where historical context matters, code reviews, and onboarding new contributors to existing Zed projects.

3. GitHub MCP Server — Issues, PRs, and Code Search

The GitHub MCP server takes Git context to the platform level. Your AI assistant can browse issues, review PRs, search across all your repositories, and manage GitHub workflows — all without leaving your AI chat context. This pairs well with Zed's collaborative editing for teams working on GitHub-hosted projects.

Key capabilities:

  • Browse and create issues and pull requests
  • Search code across all repositories in your organization
  • Review PR diffs and add comments
  • Access repository metadata, tags, and releases

Best for: Teams using GitHub for collaboration alongside Zed. Gives your AI assistant the full project management context alongside the local development context.

4. PostgreSQL MCP Server — Real Database Context

When your Zed project connects to a PostgreSQL database, the PostgreSQL MCP server gives your AI assistant the actual schema — tables, columns, foreign keys, and constraints — rather than forcing it to guess from code. Queries, migrations, and ORM configurations generated with real schema context are accurate; those generated without it often aren't.

Key capabilities:

  • Introspect live database schema (tables, types, relationships)
  • Run read-only queries for data validation and debugging
  • Generate accurate migrations based on real schema state
  • Support multiple database connections for multi-tenant projects

Best for: Backend development in Zed where database accuracy matters. Essential for any project using PostgreSQL, Supabase, or Neon.

5. Brave Search MCP Server — Documentation and Error Lookup

Zed's AI assistant has a training cutoff. The Brave Search MCP server closes that gap — giving your AI real-time access to documentation, Stack Overflow answers, GitHub issues, and release notes. When you hit a library error or need the latest API reference, your AI can look it up rather than hallucinating from stale training data.

Key capabilities:

  • Web search for current documentation and error messages
  • Look up library changelogs and breaking changes
  • Find real-world implementation examples
  • Research unfamiliar APIs with current reference material

Best for: Any Zed development session involving third-party libraries, unfamiliar APIs, or recent framework updates. Eliminates the hallucination risk that comes from stale training data.

The Zed Developer MCP Stack

  • Local files: Filesystem MCP (full project navigation)
  • Version control: Git MCP (history, blame, diffs)
  • Platform: GitHub MCP (issues, PRs, code search)
  • Database: PostgreSQL MCP (live schema access)
  • Research: Brave Search MCP (real-time documentation)

Zed ships with excellent AI integration at the editor level. MCP servers complete the picture at the context level — giving your AI assistant structured, real-time access to everything that exists outside the current file. The combination produces an AI development environment where nothing gets lost in translation between what you know about your project and what your AI assistant can reason about.

Related guides:

Recommended Tools

Better Stack

Free Plan

Get alerted when your APIs, browser tests, payment pipelines, or MCP server dependencies go down. Used by 100K+ developers.

Start monitoring free →

1Password

14-day Free Trial

Store and inject API keys, payment credentials, tokens, and file access secrets into your MCP server configs. Trusted by 150K+ developers.

Try 1Password free →

🔧 MCP Servers Mentioned in This Article

📁

Filesystem MCP Server

sandboxed read, write, edit, move and search access to an explicit whitelist of local directories, and it is the reference implementation most other filesystem MCP servers are modelled on. Shipped by Anthropic in the official modelcontextprotocol/servers monorepo (89,000+ stars, actively maintained), it is a Node.js server published to npm as @modelcontextprotocol/server-filesystem. The part worth understanding before you install is the access-control model, because there are now two ways to grant directories and they do not compose. Method one is command-line arguments: `npx -y @modelcontextprotocol/server-filesystem /path/one /path/two`. Method two, and the one the maintainers recommend, is MCP Roots — a client that supports the roots protocol sends its roots at initialization, and those roots COMPLETELY REPLACE any directories passed on the command line, then get replaced again on every `notifications/roots/list_changed`. That means allowed directories can change at runtime without restarting the server, but it also means a roots-capable client silently overrides your CLI arguments. If the server starts with no arguments and the client either does not support roots or sends an empty list, initialization throws an error. The tool surface is broad: `read_text_file` (with mutually exclusive `head`/`tail` line windows), `read_media_file` returning base64 image/audio content blocks, `read_multiple_files` which keeps going when individual reads fail, `write_file`, `edit_file`, `create_directory`, `list_directory`, `list_directory_with_sizes`, `move_file`, `search_files`, `directory_tree`, `get_file_info` and `list_allowed_directories`. `edit_file` is the one to learn — it does line-based and multi-line pattern matching with indentation detection and preservation, returns a git-style diff with context, and supports `dryRun: true` so you can preview a change before applying it; the maintainers recommend always running a dry run first. Every operation is refused outside the allowed set, and `list_allowed_directories` is the fastest way to confirm what the server actually believes it can touch.

Local
💻

GitHub MCP Server

authenticated access to the whole GitHub platform — repositories, files, branches, issues, pull requests, Actions runs, security alerts, discussions and notifications — from Claude, Cursor, VS Code, Copilot CLI and any other MCP host. There is no npm package for this server, and that trips up most people who try to install it: `@github/mcp-server` is not published to the npm registry, so any `npx` line you find for it will fail. GitHub ships it three other ways. The easiest is the hosted remote server at https://api.githubcopilot.com/mcp/, which needs no install at all — point an HTTP-transport MCP client at that URL and log in with OAuth (VS Code 1.101+, Claude Desktop, Claude Code, Cursor and Windsurf all support this). The second is the official Docker image ghcr.io/github/github-mcp-server, which is what the copy-paste command on this page runs; on github.com it now performs a browser-based OAuth login on first use and keeps the token in memory only, which is why the published Docker configs map a fixed loopback callback port (-p 127.0.0.1:8085:8085 with GITHUB_OAUTH_CALLBACK_PORT=8085) so the container can receive the callback. Prefer a token? Set GITHUB_PERSONAL_ACCESS_TOKEN instead — it takes precedence over OAuth, and the minimum useful scopes are repo, read:org and read:packages. The third is the native Go binary from the repository's releases, which needs no fixed port for the OAuth flow. GitHub Enterprise Server has no hosted option: use the local server with --gh-host or GITHUB_HOST set to your instance (include the https:// scheme — it defaults to http://, which GHES rejects). Toolsets can be narrowed with GITHUB_TOOLSETS, and an insiders channel is available at /mcp/insiders or via the X-MCP-Insiders header.

Auth required📘
🔍

Brave Search MCP Server

The Brave Search MCP Server is the official server from Brave that gives AI assistants privacy-first web search through the independent Brave Search API — no tracking, no profiling, and results drawn from Brave's own web index rather than Google or Bing. It exposes five distinct tools that map directly to the Brave Search API endpoints: brave_web_search for general queries with pagination, freshness filters, and safe-search controls; brave_local_search for businesses, restaurants, and points of interest with automatic location filtering; brave_news_search for recent articles and current events; brave_image_search for image discovery; and brave_video_search for finding videos across the web. Authentication uses a single BRAVE_API_KEY (free tier available at brave.com/search/api) or a mounted BRAVE_API_KEY_FILE for Docker-secret setups. Install in Claude Desktop, Cursor, Windsurf, or VS Code with one npx command and choose stdio or streamable-HTTP transport. Because Brave operates its own crawler and index, the Brave Search MCP server is a strong choice for developers who want an alternative to Google-dependent search tools, need reproducible non-personalized results, or care about data privacy in agent workflows — Claude can pull fresh web context, verify facts, and research topics without leaking queries to ad-tech pipelines.

Local
💻

Git

Tools to read, search, and manipulate Git repositories. Full Git operations support.

Local
🗄️

PostgreSQL MCP Server

The PostgreSQL MCP server was the Model Context Protocol reference server for Postgres, and it is retired: the source now sits in modelcontextprotocol/servers-archived — a repository GitHub reports as archived, described as "Reference MCP servers that are no longer maintained" — and the npm package @modelcontextprotocol/server-postgres carries a deprecation notice reading "Package no longer supported." It still installs and still runs, which is why most third-party setup articles have not caught up. What it provides is deliberately small: a single tool, query, which executes read-only SQL inside a READ ONLY transaction, plus per-table schema information exposed as MCP resources at postgres://<host>/<table>/schema, with column names and data types discovered from database metadata. There is no index advice, no health check, no separate schema-listing tool, and no write mode. Install is npx @modelcontextprotocol/server-postgres with a postgres:// connection string as the argument. For active work against Postgres, the maintained alternative is Postgres MCP Pro (crystaldba/postgres-mcp), which exposes nine tools including index tuning against hypothetical indexes and a database health check, and has an explicit restricted access mode; if your database is hosted on Supabase or Neon, their platform servers add branching and logs that a raw Postgres connection cannot see. Reach for this archived server only when you want the smallest possible surface — one process, one read-only query tool, nothing else.

Local📘

📚 More from the Blog