Get the FREE Ultimate OpenClaw Setup Guide →

linear-streamable

MCP Server for interacting with Linear API. Written in TypeScript, Node and Hono.dev

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio iceener-linear-streamable-mcp-server bun dev \
  --env PORT="3000" \
  --env BEARER_TOKEN="<Your Linear API key>" \
  --env AUTH_STRATEGY="bearer"

How to use

This MCP server provides a streamable, LLM-friendly interface for Linear's workspace data and actions. It aggregates common Linear entities (issues, projects, teams, users, cycles) and exposes batch-enabled operations so an AI agent can perform multiple steps in a single call. The toolset focuses on discoverability (workspace_metadata), searching and filtering (list_issues), and batch actions (e.g., create_issues) while returning human-readable feedback and diffs to help users verify changes. You can run the server locally or deploy it as a Cloudflare Worker for remote access, and connect your MCP client to the provided http endpoint. Tools are designed to be discoverable and self-describing, reducing API jargon and improving the agent’s ability to reason about IDs and relationships within the Linear workspace.

How to install

Prerequisites:

  • Bun (https://bun.sh/) or Node.js 24+ with Bun-compatible setup
  • Linear account to acquire an API key
  • Optional: Cloudflare account for remote/edge deployment

Installation steps:

  1. Clone the repository and install dependencies
git clone <repo-url>
c d linear-mcp
bun install
  1. Create or copy the environment configuration
cp env.example .env
  1. Edit the environment file with your configuration, notably:
PORT=3000
AUTH_STRATEGY=bearer
BEARER_TOKEN=lin_api_xxxx  # Your Linear API key
  1. Start the server locally
bun dev
# MCP: http://127.0.0.1:3000/mcp
  1. (Optional) If you want to test via an MCP client, use the client config example in the README under Client Configuration.

Additional notes

Tips and notes:

  • The server supports two runtimes: Node/Hono locally and Cloudflare Worker for remote access. For production deployments, consider securing tokens, TLS, CORS, rate limiting, and audit logging.
  • Environment variables control authentication and exposure. Replace placeholder values with real tokens and secrets in a secure manner.
  • The tools are batch-oriented where available (e.g., create_issues accepts arrays). Use workspace_metadata first to fetch IDs (teams, projects, states, labels) to minimize API chatter.
  • If you run into token or RS-minted token issues, check AUTH_STRATEGY and related RS settings in your environment. The example configuration uses a Bearer strategy for local development.
  • Cloudflare deployment steps are described in the README; ensure you update wrangler config and secrets if you go remote.

Related MCP Servers

Sponsor this space

Reach thousands of developers