Linear MCP Server
Streamable HTTP MCP server for Linear — manage issues, projects, teams, cycles, and comments.
Author: overment
[!WARNING]
You connect this server to your MCP client at your own responsibility. Language models can make mistakes, misinterpret instructions, or perform unintended actions. Review tool outputs, verify changes (e.g., with list_issues), and prefer small, incremental writes.
The HTTP/OAuth layer is designed for convenience during development, not production-grade security. If deploying remotely, harden it: proper token validation, secure storage, TLS termination, strict CORS/origin checks, rate limiting, audit logging, and compliance with Linear's terms.
Comparison
Below is a comparison between the official Linear MCP (top) and this MCP (bottom).
Notice
This repo works in two ways:
- As a Node/Hono server for local workflows
- As a Cloudflare Worker for remote interactions
For production Cloudflare deployments, see Remote Model Context Protocol servers (MCP).
Motivation
I'm a big fan of Linear and use it daily. At the time of writing, the official MCP server isn't fully optimized for language models. This server is built with key goals in mind:
- Let LLMs find Team IDs, Project IDs, Status IDs, or User IDs in a single action (
workspace_metadata) instead of multiple tool calls
- Include clear MCP instructions and schema descriptions that cut API jargon
- Map API responses into human-readable feedback — useful for both the LLM and user
- Provide hints and suggestions for next steps, plus tips on recovering from errors
- Support batch actions (e.g.,
create_issues instead of create_issue) so the LLM can perform multiple steps in one go
- Prefetch related values — return both a status ID and actual status name for an issue
- Hide tools not enabled in a given team's settings (like
list_cycles) to reduce noise
In short, it's not a direct mirror of Linear's API — it's tailored so AI agents know exactly how to use it effectively.
Features
- ✅ Issues — List, search, create, update (state, assignee, labels, priority, etc.)
- ✅ Projects — List, create, update projects
- ✅ Teams & Users — Discover workspace structure
- ✅ Cycles — Browse sprint/cycle planning
- ✅ Comments — List and add comments on issues
- ✅ OAuth 2.1 — Secure PKCE flow with RS token mapping
- ✅ Dual Runtime — Node.js/Bun or Cloudflare Workers
- ✅ Production Ready — Encrypted token storage, rate limiting, multi-user support
Design Principles
- LLM-friendly: Tools are simplified and unified, not 1:1 API mirrors
- Batch-first: Create/update operations accept arrays to minimize tool calls
- Discovery-first:
workspace_metadata returns all IDs needed for subsequent calls
- Clear feedback: Every response includes human-readable summaries with diffs
Installation
Prerequisites: Bun, Node.js 24+, Linear account. For remote: a Cloudflare account.
Ways to Run (Pick One)
- Local (API key) — Fastest start
- Local + OAuth — For multi-user or token refresh
- Cloudflare Worker (wrangler dev) — Local Worker testing
- Cloudflare Worker (deploy) — Remote production
1. Local (API Key) — Quick Start
Run the server with your Linear Personal Access Token from Settings → Security.
git clone <repo>
cd linear-mcp
bun install
cp env.example .env
Edit .env:
PORT=3000
AUTH_STRATEGY=bearer
BEARER_TOKEN=lin_api_xxxx # Your Linear API key
bun dev
# MCP: http://127.0.0.1:3000/mcp
Connect to your MCP client:
Claude Desktop / Cursor:
{
"mcpServers":
... [View full README on GitHub](https://github.com/iceener/linear-streamable-mcp-server#readme)
