Menu
Mattering Automations connector

The Linear MCP for teams that want issue tracking to run itself

Connect Linear once and your issues, comments, projects, milestones, cycles, labels, and workflow states become building blocks for visual automations and AI agent workflows. 35 nodes over Linear's GraphQL API, including a catch-all node for anything else in the schema.

35 ready-made nodes One credential to connect Works with Claude, ChatGPT, and Matter

What the Linear MCP connector does

The Linear connector for Mattering Automations covers the part of Linear a product team touches every day. Create, update, search, archive, unarchive, and delete issues, add and remove labels, post and edit comments, create and update projects, read project milestones and cycles, manage issue labels, and resolve the teams, users, and workflow states that every other call depends on. That is 35 nodes: 18 read nodes and 17 action nodes.

Linear has no REST API. Its public API is a single GraphQL endpoint at api.linear.app/graphql, the same one the Linear app runs on, so every typed node here is a named GraphQL operation with a fixed document behind it. You fill in ordinary canvas fields such as a team id or a title, the node maps them onto that operation's variables, and nothing asks you to write GraphQL. A fixed document is also why the fields are honest: only arguments that exist in Linear's published schema are on the node.

Each capability is a node on the Automations canvas, and every node speaks MCP, the open protocol for connecting AI assistants to tools. Claude, ChatGPT, and any MCP-capable assistant can use them, and they also power Matter, the platform's built-in AI assistant. Describe the issue workflow you want in plain language and it gets built on the canvas in front of you, where you can see every step and change it.

The list nodes are deliberately lean: they return the rows and leave nested collections such as labels and comments out, because Linear caps a single query at 10,000 complexity points and a page that expanded every relationship would spend that budget on data most runs never read. When you need the whole record, Linear: Get Issue returns it. Past the typed nodes, Linear: Custom GraphQL Request runs any Linear GraphQL operation with your own document and variables, so parts of the schema without a typed node stay reachable.

One connection, three surfaces

  • Visual automations. Drag Linear nodes onto the canvas and wire them to triggers, conditions, and other apps.
  • AI agents, yours or ours. These nodes speak MCP, the open protocol, so Claude, ChatGPT, and any MCP-capable assistant can use them, and Matter, the agent built into the platform, uses them natively.
  • The whole platform. Combine with native tasks, projects, and other connectors, no middleware in between.

Every Linear node, ready to drop on the canvas

35 nodes, generated from the connector itself so this list always matches what ships. Expand a category to browse.

Read and query nodes 18

Linear: Get Current User

Get the Linear account the API key belongs to. Takes no parameters, so it is the cheapest auth and connectivity probe.

Linear: Get Issue

Get one issue with its full record: description, state, assignee, team, project, cycle, parent, and labels.

Linear: Get Organization

Get the Linear workspace (organization) the API key belongs to, including its URL key and seat count.

Linear: Get Project

Get one project with its full body content, its teams, and its members.

Linear: Get Team

Get one team with its workflow states, its members, and its active cycle. The fastest way to resolve the state id and assignee id you need to file an issue.

Linear: Get User

Get one Linear user by id.

Linear: Get Workflow State

Get one workflow state by id, including which team it belongs to.

Linear: List Cycles

List cycles (sprints) with their dates and progress. Filter by team, then use the cycle id when creating or moving issues.

Linear: List Issue Comments

List the comments on an issue, including who wrote each one and whether it has been resolved.

Linear: List Issue Labels

List issue labels. Workspace labels have a null team; team labels carry their team. Use it to resolve label ids for issue creation.

Linear: List Issues

List issues across the workspace with a structured filter, e.g. every started issue on one team. Descriptions and labels are left out to keep the query cheap; use Linear: Get Issue for the full record.

Linear: List Project Issues

List the issues attached to a project, optionally narrowed with an issue filter.

Linear: List Project Milestones

List project milestones, optionally filtered to one project. Use it to resolve the milestone id an issue should be filed against.

Linear: List Projects

List projects in the workspace with their status, lead, dates, and progress.

Linear: List Teams

List the teams in the workspace with their keys. Every issue belongs to a team, so this is usually the first call in a Linear automation.

Linear: List Users

List the members of the Linear workspace. Use it to resolve a user id for an assignee before creating or updating an issue.

Linear: List Workflow States

List workflow states (Backlog, Todo, In Progress, Done, Canceled) across the workspace. Filter by team to get the state ids that Linear: Update Issue needs.

Linear: Search Issues

Full text search across issues. Use this when you have words rather than ids; use Linear: List Issues when you have a structured filter.

Action nodes 17

Linear: Add Label to Issue

Add one label to an issue without touching its other labels.

Linear: Archive Issue

Archive an issue, optionally moving it to the trash. Reversible with Linear: Unarchive Issue.

Linear: Create Comment

Post a comment on an issue, a project, or a project update. Supply exactly one target id.

Linear: Create Issue

Create a Linear issue. Only the team id and a title are required; everything else is optional.

Linear: Create Issue Label

Create an issue label. Supply a team id for a team label, or leave it blank for a workspace label.

Linear: Create Project

Create a project. Linear requires a name and at least one owning team.

Linear: Custom GraphQL Request

Advanced: run any Linear GraphQL operation with your own document and variables. Covers everything not individually typed, including initiatives, documents, attachments, customers, and admin mutations.

Linear: Delete Comment

Delete a comment. Destructive and not recoverable through the API.

Linear: Delete Issue

Delete an issue. By default it goes to the trash; set permanently_delete to remove it outright. Destructive.

Linear: Delete Issue Label

Delete an issue label and remove it from every issue that carries it. Destructive.

Linear: Delete Project

Delete a project. Its issues stay in their teams but lose the project link. Destructive.

Linear: Remove Label from Issue

Remove one label from an issue. The label itself is not deleted.

Linear: Unarchive Issue

Restore an archived issue back into its team's active issue list.

Linear: Update Comment

Edit a comment's body, or mark it resolved by naming the resolving user.

Linear: Update Issue

Update an issue: move it to another state, reassign it, reprioritise it, retitle it, or move it between teams and projects. Only the fields you fill in are sent.

Linear: Update Issue Label

Rename an issue label or change its colour, description, or parent group.

Linear: Update Project

Update a project: rename it, move its status, change its lead, or shift its dates. Only the fields you fill in are sent.

Connecting Linear takes one credential

Linear authenticates with a personal API key that you create in your Linear workspace under Settings, then API, where the keys live under Personal API keys. The key carries your own permissions across every team you can see, so create it from an account whose access matches what you want automations to reach. One detail matters more here than on most connectors: the key is sent as a raw Authorization header with no Bearer prefix. Bearer is only correct for OAuth access tokens, and Linear rejects an API key sent that way.

1

In Linear, open Settings, then API, and create a key under Personal API keys. Give it a label that says where it will be used, such as Mattering Automations.

2

Copy the key. Paste it exactly as Linear shows it, with no Bearer prefix and no quotes around it. Mattering strips a pasted Bearer prefix defensively, but the raw key is the correct value.

3

In Mattering Automations, open Connectors, click Add Connector, and choose Linear.

4

Create a new credential, paste the key, and save. Credentials are stored encrypted in Google Secret Manager and never appear in workflow logs.

5

Connect the connector, then drop Linear: Get Current User onto a canvas and run it. It takes no parameters and returns the account the key belongs to, which confirms the credential works. All 35 Linear nodes are now in your palette.

Common uses for the Linear MCP

File issues from wherever the work actually arrives

A support email, a form submission, a failed deploy, or an agent instruction becomes a Linear issue on the right team with a title, description, assignee, priority, and labels already set. Linear: List Teams and Linear: List Workflow States resolve the ids first, so the issue lands filed rather than in an inbox someone has to sort.

Triage the backlog with an AI step in the middle

Pull untriaged issues with Linear: List Issues or Linear: Search Issues, hand each one to an AI step to judge severity and area, then apply the decision with Linear: Update Issue and Linear: Add Label to Issue. The judgement is a step on the canvas you can read and change, not a black box.

Report on a cycle without anyone assembling it

Linear: List Cycles returns the active sprint and its progress, Linear: List Issues narrows to that cycle and team, and the result becomes a written summary posted to a channel, an email, or a task every Friday afternoon.

Keep projects and milestones honest

Read a project with Linear: Get Project, list its open work with Linear: List Project Issues, check what is still hanging off a milestone with Linear: List Project Milestones, and move the project's status or dates with Linear: Update Project when the work says something different from the plan.

Mirror Linear into the rest of your workspace

Turn a new Linear issue into a tracked task in Mattering Productivity, or push an issue's state change into a customer-facing update, so the people who do not live in Linear stop asking where something stands.

Post context back onto the issue itself

Linear: Create Comment writes onto an issue, a project, or a project update, so the log excerpt, the customer reply, or the agent's summary lands where the engineer is already looking instead of in a separate system.

Example: An inbound bug report becomes a triaged, labelled Linear issue

Built on the Automations canvas in a few minutes, or described in plain language to Matter and built for you.

1

Trigger: a form submission, an inbound email, or a schedule that checks a queue.

2

An AI step reads the report and drafts a clear title, a reproduction summary, and a severity.

3

Linear: List Teams resolves the team id, and Linear: List Workflow States returns the state ids for that team.

4

Linear: Search Issues checks for an existing issue about the same thing, and a condition node branches on whether one was found.

5

Linear: Create Issue files the new issue with the drafted title and description, the team, the priority, and an assignee, or Linear: Create Comment adds the new report to the duplicate that already exists.

6

Linear: Add Label to Issue tags it so the backlog stays sortable, and Mattering Productivity gets a task carrying the issue identifier for anyone tracking the work outside Linear.

Linear MCP questions, answered

Can an AI agent create, update, and triage Linear issues through this MCP?

Yes. Linear: Create Issue files an issue, Linear: Update Issue moves it between states, reassigns it, changes its priority, or moves it to another team or project, and Linear: Add Label to Issue and Linear: Remove Label from Issue handle labelling. Linear: Archive Issue, Linear: Unarchive Issue, and Linear: Delete Issue cover the end of an issue's life. Matter and any MCP-connected assistant can call them, and so can a scheduled or event-driven workflow on the canvas. Delete is destructive, so it is worth putting a condition or an approval step in front of it.

Does this work with Claude, ChatGPT, or other AI assistants?

Yes. Mattering exposes these nodes over MCP, an open protocol for connecting AI assistants to tools, so any MCP-capable assistant, including Claude and ChatGPT, can connect to your workspace and work with Linear through them. Matter, the assistant built into the platform, uses the same nodes natively.

Linear's API is GraphQL. Do I need to know GraphQL to use these nodes?

No. Each of the 34 typed nodes carries a fixed GraphQL operation and maps your canvas fields onto its variables, so filing an issue means typing a title and picking a team id exactly as you would on any other connector. GraphQL only becomes visible if you choose Linear: Custom GraphQL Request, the advanced node that takes a document and variables of your own.

What credential does the connector use, and why is there no Bearer prefix?

A Linear personal API key, created under Settings, then API, then Personal API keys, and stored encrypted in Google Secret Manager. Linear expects that key raw in the Authorization header. The Bearer prefix belongs to OAuth access tokens only, and Linear's docs show both samples on the same page, so copying the OAuth one is the most common way a Linear setup fails. Paste the key on its own. Mattering also strips a leading Bearer if one slips in, rather than letting every node fail with an opaque auth error.

Can a Linear event start a workflow?

Not yet. This connector ships read and action nodes only, with no trigger node, so a Linear event cannot currently be the first step of an automation. The usual pattern in the meantime is a scheduled workflow that polls with Linear: List Issues or Linear: Search Issues and acts on what changed. Linear does support outbound webhooks, so a proper trigger is a design question rather than a missing capability in Linear.

What are Linear's API rate limits?

Linear enforces hourly limits on both request count and query complexity, and they differ by how you authenticate and which plan the workspace is on. Linear's own rate limiting documentation states two different request-per-hour figures for API keys in different places, so we do not repeat a number here: check their docs for the current one. The cap worth designing around is the per-query one, because a single GraphQL query is limited to 10,000 complexity points. The connector does not back off for you, so pace loops that call list nodes repeatedly.

Why do the list nodes leave out labels and comments on each row?

Because of that complexity cap. A 50-issue page that also expanded labels, comments, and subscribers per row would spend a large share of a single query's 10,000 point budget on data most runs never read, and a big enough page would simply fail. The list nodes return the rows, Linear: Get Issue returns one issue's full record including its labels, and Linear: List Issue Comments returns the comments when you actually want them.

What is not covered by the typed nodes?

Initiatives, documents, attachments, customers, and roadmaps do not have typed nodes: they are real parts of Linear's schema but sit outside the issues, comments, projects, teams, users, labels, and states brief this connector was built to. Linear's batch issue create and update operations were also left out, because bulk editing needs canvas handling we would rather build properly than half wire. All of it is reachable through Linear: Custom GraphQL Request, and every typed node also accepts an extra variables object that is merged into the GraphQL variables, so input-object fields a node does not expose individually still work.

35 nodes waiting on the canvas

Put Linear to work inside your workspace

Connect once, then use every node in visual automations and AI agent workflows across the Mattering platform.

All product names, logos, and brands are property of their respective owners and are used for identification purposes only. Use of these names, logos, and brands does not imply endorsement or affiliation. Linear is a product of its respective owner; Mattering is an independent platform.

Start Free Explore our Products