About Linear MCP Server Extension for Zed
A Zed extension that connects its AI Assistant to a Linear MCP server. It supports issue, comment, project, and workspace operations, with tool prefixes for distinguishing multiple workspaces.
Key features
- 01Create and manage issues
- 02Create, edit, delete, and resolve comments
- 03Manage projects and milestones
- 04Search and filter issues
- 05Manage labels and issue relationships
- 06Use multiple workspaces with tool prefixes
Use cases
- 01Create or update Linear issues from Zed
- 02Manage comments while working in Zed
- 03Find issues across a Linear workspace
- 04Work with multiple Linear workspaces in one Zed setup
MCP support
The extension integrates a Linear MCP server into Zed's AI Assistant, exposing Linear operations as MCP tools.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
24Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| code | string | Yes | OAuth authorization code |
Details
From the project READMELinear MCP Server Extension for Zed
This extension integrates my Linear MCP Server into Zed's AI Assistant, enabling direct interaction with Linear through Zed.

Features
- ✨ Full Linear API integration through MCP
- 🔄 Create, update, and manage issues directly from Zed
- 📝 Comment management (create, update, delete, resolve)
- 📊 Project and milestone management
- 👥 Multiple workspace support
- 🔍 Advanced issue search and filtering
- 🏷️ Label management
- 📎 Parent/child issue relationships
Setup
Option 1: Developer Token
Go to Linear: Workspace Settings > API > OAuth application > Create or select an application (e.g., "Cline MCP"). Under "Developer Token", click "Create & copy token". Select "Application" as the actor and copy the generated token.
Option 2: Personal API Key
Go to Linear: Your Personal Settings > API > Personal API Keys. Click "Create key", give it a label (e.g., "Cline MCP"), and copy the generated key.
Setting up Zed
Take the key you just received and add them to your Zed settings.json, in the "context_servers" field
{
"context_servers": {
"mcp-server-linear": {
"settings": {
"linear_api_key": "<LINEAR_API_KEY>",
"tool_prefix": "<optional tool prefix>"
}
}
}
}
You can also add these keys directly into an Agent profile, for example:
{
"assistant": {
"profiles": {
"write": {
"name": "Write",
"context_servers": {
"mcp-server-linear": {
"settings": {
"linear_api_key": "<LINEAR_API_KEY>",
"tool_prefix": "company1" // optional
}
}
}
}
}
}
}
Multiple Workspace Support
If you work with multiple Linear workspaces, you can use the optional tool_prefix setting to prefix all Linear tools. This helps distinguish between different workspaces:
{
"context_servers": {
"mcp-server-linear": {
"settings": {
"linear_api_key": "<LINEAR_API_KEY>",
"tool_prefix": "company1"
}
}
}
}
With this configuration:
- Tool names will be prefixed (e.g.,
company1_linear_create_issue) - Makes it clear which workspace each tool is operating on
- Prevents conflicts when working with multiple Linear instances
Multiple Simultaneous Installations
You can run multiple instances of the MCP server simultaneously by configuring different command blocks with unique TOOL_PREFIX values:
{
"context_servers": {
"mcp-server-linear": {
// provided through Zed extension
"settings": {
"linear_api_key": "<LINEAR_API_KEY>",
"tool_prefix": "company1"
}
},
"linear-workspace2": {
// provided through npx
"settings": {},
"command": {
"command": "npx",
"args": ["-y", "mcp-server-linear"],
"env": {
"LINEAR_ACCESS_TOKEN": "<LINEAR_API_KEY_2>",
"TOOL_PREFIX": "workspace2"
}
}
}
}
}
Each instance will have its tools prefixed with the specified TOOL_PREFIX, allowing you to interact with multiple Linear workspaces simultaneously.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.