Cara menambahkan MCP server ke Zed

Zed is a fast code editor with a built-in AI agent. MCP servers are called context servers in Zed and are added from Settings → AI → MCP Servers or in settings.json.

✓ stdio✓ HTTP? SSE✓ OAuthBerjalan di: Zed editorDiverifikasi dengan dokumentasi Zed

Tambahkan MCP server ke Zed

1Run zed: open settings file, or use Settings → AI → MCP Servers → Add Server.
2Paste the server under context_servers.
3Remote servers take url; without an Authorization header Zed runs the MCP OAuth flow.

Contoh: Codebase Memory

settings.json

{
  "context_servers": {
    "codebase-memory-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "codebase-memory-mcp"
      ]
    }
  }
}

Context servers and tool permissions

Zed calls MCP servers context servers and keeps them under context_servers in settings.
Permissions can be set per tool with keys like mcp:<server>:<tool>: confirm, allow or deny.
A server that changes its tool list is picked up without restarting Zed.
A remote server without an Authorization header triggers the standard MCP OAuth sign-in.

Lokasi MCP server di Zed

settings.jsonzed: open settings file

Cek apakah sudah dimuat: Agent panel → server status

Batasan dan pengaturan

The config key is context_servers, not mcpServers.
Zed supports MCP tools and prompts.
Tool permissions default to confirm; set agent.tool_permissions per tool with keys like mcp:<server>:<tool>.
Servers can update their tool list without a restart.

Saat server tidak berjalan di Zed

A green dot next to the server name means it is active.
Check the server's own logs or docs for the error details.
Naming the server in your prompt helps the model pick its tools.

Pertanyaan seputar MCP di Zed

How do I add an MCP server to Zed?

Open Settings → AI → MCP Servers → Add Server, or add an entry under context_servers in your settings file.

Does Zed support remote MCP servers?

Yes. A remote server takes a url, and Zed starts the OAuth flow if no Authorization header is set.

Why is the key context_servers?

That is Zed's name for MCP servers in its settings; the entries work the same way.

Can I allow a tool without confirming each time?

Yes, with agent.tool_permissions set to allow for that tool.

MCP server populer untuk Zed

1
Codebase Memorydeusdata/codebase-memory-mcp48,6 rb stars
Provides a local knowledge graph that connects source code, documentation, and infrastructure to give AI agents structural context.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "codebase-memory-mcp": { "command": "npx", "args": [ "-y", "codebase-memory-mcp" ] } } }
2
LLM Sandboxvndee/llm-sandbox1,1 rb stars
Securely execute LLM-generated code in isolated containers or local Linux and Windows sandboxes across multiple programming languages.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "llm-sandbox": { "command": "uvx", "args": [ "llm-sandbox" ] } } }
3
Pascal 3D Editorpascalorg/editor25,5 rb stars
Lets an AI client create, inspect, validate, and save editable 3D building scenes, including walls, rooms, doors, and furnishings.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "editor": { "command": "npx", "args": [ "-y", "@pascal-app/mcp" ] } } }
4
Apify MCP Serverapify/apify-mcp-server16 rb stars
Lets AI agents search, run, and read results from thousands of web scraping and automation Actors on Apify Store.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "apify-mcp-server": { "url": "https://mcp.apify.com/" } } }
5
CocoIndex Codecocoindex-io/cocoindex-code3 rb stars
Provides AST-based semantic code search for local repositories using incremental re-indexing and natural language queries.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "cocoindex-code": { "command": "uvx", "args": [ "cocoindex-code" ] } } }
6
AdLoopklosk/adloop301 stars
Lets an AI client read and change Google Ads, Reddit Ads and GA4 campaigns, with a preview shown before every change.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "adloop": { "command": "uvx", "args": [ "adloop" ] } } }
7
Claude Flowruvnet/claude-flow81,6 rb stars
An agent meta-harness that provides Claude Code and Codex with specialized agent swarms, self-learning memory, and enterprise security guardrails.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "claude-flow": { "command": "npx", "args": [ "-y", "claude-flow" ] } } }
8
Playwrightmicrosoft/playwright-mcp39,6 rb stars
Enables LLMs to interact with web pages through structured accessibility snapshots using Playwright browser automation.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "playwright-mcp": { "command": "npx", "args": [ "-y", "@playwright/mcp" ] } } }
9
Atlassian Dc MCP Bitbucketb1ff/atlassian-dc-mcp109 stars
Allows an AI client to interact with Bitbucket Data Center repositories and code through the Model Context Protocol.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "atlassian-dc-mcp": { "command": "npx", "args": [ "-y", "@atlassian-dc-mcp/bitbucket" ] } } }
10
Next AI Draw.iodayuanjiang/next-ai-drawio36,9 rb stars
Lets an AI agent draw and edit draw.io diagrams on a live browser canvas that you can also edit by hand.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "next-ai-drawio": { "command": "npx", "args": [ "-y", "@next-ai-drawio/mcp-server" ] } } }
11
Repomixyamadashy/repomix29,3 rb stars
Packs a local or remote codebase into a single AI-friendly file with token counts so LLMs and coding agents can read or search it.
Konfigurasi untuk Zed
settings.json
{ "context_servers": { "repomix": { "command": "npx", "args": [ "-y", "repomix" ] } } }

MCP Toolbox for Databasesgoogleapis/mcp-toolbox16,9 rb stars

Connects AI agents and IDEs to enterprise databases using prebuilt tools or a custom framework for secure, structured queries.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "mcp-toolbox": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}

Tambahkan MCP server ke Zed

1Run zed: open settings file, or use Settings → AI → MCP Servers → Add Server.

2Paste the server under context_servers.

3Remote servers take url; without an Authorization header Zed runs the MCP OAuth flow.

Contoh: Codebase Memory

settings.json

{
  "context_servers": {
    "codebase-memory-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "codebase-memory-mcp"
      ]
    }
  }
}

Context servers and tool permissions

Zed calls MCP servers context servers and keeps them under context_servers in settings.

Permissions can be set per tool with keys like mcp:<server>:<tool>: confirm, allow or deny.

A server that changes its tool list is picked up without restarting Zed.

A remote server without an Authorization header triggers the standard MCP OAuth sign-in.

Pertanyaan seputar MCP di Zed

How do I add an MCP server to Zed?

Open Settings → AI → MCP Servers → Add Server, or add an entry under context_servers in your settings file.

Does Zed support remote MCP servers?

Yes. A remote server takes a url, and Zed starts the OAuth flow if no Authorization header is set.

Why is the key context_servers?

That is Zed's name for MCP servers in its settings; the entries work the same way.

Can I allow a tool without confirming each time?

Yes, with agent.tool_permissions set to allow for that tool.

MCP server populer untuk Zed

Codebase Memorydeusdata/codebase-memory-mcp48,6 rb stars

Provides a local knowledge graph that connects source code, documentation, and infrastructure to give AI agents structural context.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "codebase-memory-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "codebase-memory-mcp"
      ]
    }
  }
}

LLM Sandboxvndee/llm-sandbox1,1 rb stars

Securely execute LLM-generated code in isolated containers or local Linux and Windows sandboxes across multiple programming languages.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "llm-sandbox": {
      "command": "uvx",
      "args": [
        "llm-sandbox"
      ]
    }
  }
}

Pascal 3D Editorpascalorg/editor25,5 rb stars

Lets an AI client create, inspect, validate, and save editable 3D building scenes, including walls, rooms, doors, and furnishings.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "editor": {
      "command": "npx",
      "args": [
        "-y",
        "@pascal-app/mcp"
      ]
    }
  }
}

Apify MCP Serverapify/apify-mcp-server16 rb stars

Lets AI agents search, run, and read results from thousands of web scraping and automation Actors on Apify Store.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "apify-mcp-server": {
      "url": "https://mcp.apify.com/"
    }
  }
}

CocoIndex Codecocoindex-io/cocoindex-code3 rb stars

Provides AST-based semantic code search for local repositories using incremental re-indexing and natural language queries.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "cocoindex-code": {
      "command": "uvx",
      "args": [
        "cocoindex-code"
      ]
    }
  }
}

AdLoopklosk/adloop301 stars

Lets an AI client read and change Google Ads, Reddit Ads and GA4 campaigns, with a preview shown before every change.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "adloop": {
      "command": "uvx",
      "args": [
        "adloop"
      ]
    }
  }
}

Claude Flowruvnet/claude-flow81,6 rb stars

An agent meta-harness that provides Claude Code and Codex with specialized agent swarms, self-learning memory, and enterprise security guardrails.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "claude-flow": {
      "command": "npx",
      "args": [
        "-y",
        "claude-flow"
      ]
    }
  }
}

Playwrightmicrosoft/playwright-mcp39,6 rb stars

Enables LLMs to interact with web pages through structured accessibility snapshots using Playwright browser automation.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "playwright-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@playwright/mcp"
      ]
    }
  }
}

Atlassian Dc MCP Bitbucketb1ff/atlassian-dc-mcp109 stars

Allows an AI client to interact with Bitbucket Data Center repositories and code through the Model Context Protocol.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "atlassian-dc-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@atlassian-dc-mcp/bitbucket"
      ]
    }
  }
}

Next AI Draw.iodayuanjiang/next-ai-drawio36,9 rb stars

Lets an AI agent draw and edit draw.io diagrams on a live browser canvas that you can also edit by hand.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "next-ai-drawio": {
      "command": "npx",
      "args": [
        "-y",
        "@next-ai-drawio/mcp-server"
      ]
    }
  }
}

Repomixyamadashy/repomix29,3 rb stars

Packs a local or remote codebase into a single AI-friendly file with token counts so LLMs and coding agents can read or search it.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "repomix": {
      "command": "npx",
      "args": [
        "-y",
        "repomix"
      ]
    }
  }
}

MCP Toolbox for Databasesgoogleapis/mcp-toolbox16,9 rb stars

Connects AI agents and IDEs to enterprise databases using prebuilt tools or a custom framework for secure, structured queries.

Konfigurasi untuk Zed

settings.json

{
  "context_servers": {
    "mcp-toolbox": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}