Linear MCP Server is an open-source Model Context Protocol server that connects the Linear issue tracker to MCP-capable AI assistants. Published on GitHub under the MIT license with about 345 stars, it lets large language models create, search, and update Linear issues through the Linear API. Note that this is a distinct community project and not any other similarly named product that may be listed on Tooloso.
The server wraps Linear’s GraphQL API in MCP tools, so an assistant can file an issue with a title, team, description, priority, and status, update existing issues, and search with flexible filtering, all from a conversation. It runs through npx and authenticates with a Linear API key you create for your team, and it can be installed automatically for Claude Desktop through Smithery. Readers should know the repository README now marks this server as deprecated and unmaintained, pointing users to Linear’s official remote MCP server instead, so this entry mainly documents the community implementation that popularized the pattern.
Setup takes only a few minutes: create a Linear API key in your team settings, add the server block to your MCP client’s configuration file with the key in the environment, and restart the client. From there, issue management happens in the same conversation where the assistant is already helping with code, planning, and triage, instead of in a separate browser tab.
Linear MCP Server is free and open source under the MIT license. You run it yourself, and the only account involved is your own Linear workspace.
It suits engineering teams on Linear who experiment with MCP tooling and want assistants to triage and update issues conversationally, while keeping in mind the official Linear server is now the maintained option.