Linear Agent

Overview

Linear Agent turns your workspace context into action. It lives inside Linear, understands your issues, projects, teams, and history, and can act on your behalf through conversation.

You can ask it to:

• Create and update issues, projects, milestones and initiatives
• Summarize and analyze ongoing work, threads, and customer requests
• Answer questions about your workspace data
• Post, edit, and delete its own comments in threads

Reach it by opening an agent chat, using ⌘/Ctrl + J, or by mentioning @Linear in any comment within your workspace.

Setup

The Linear Agent is available by default in your workspace.

If you’d prefer to disable the Linear Agent for your workspace, a workspace admin or owner can navigate into Settings → AI → Linear Agent and toggle-off Enable Linear Agent.

Linear Agent

Linear Agent works with your existing workspace data, including teams and sub-teams, initiatives and projects (with milestones and cycles), issues and their relationships, comments, activity history, and documents.

It uses this context to understand how work is organized and to answer questions, summarize activity, and make updates in the right places.

Linear Agent operates within your existing permissions — it can only reference or change content that you already have access to.

Linear agent chat

The dedicated Linear chat is the main place to converse with Linear. It can keep context over time, show its progress as it works, and help you iterate from questions to drafts to concrete next steps.

Start a new Linear chat whenever you want a clean slate: when you’re switching to a different topic, a different issue or project scope, or you want to avoid the agent reusing earlier context from a prior thread.

If you prefer to multi-task, Linear supports working in multiple open chats at once. In the toolbar, each open chat appears as a tab so you can switch between topics without losing context. Each tab shows a short label and indicates when there are unread updates or when the agent is currently working in that thread.

For example, try prompting Linear to:

Summarize the current status of this project with progress, risks, and next steps.

Summarize the progress of our team's current cycle; what issues are at risk of rolling over?

Draft a new document on this project that comprehensively outlines the feature.

Publish a stakeholder update that is concise and focused on outcomes and risks.

You can start a new chat or return to recent conversations by typing ⌘J.

Chat history

Linear keeps a history of your past chats so you can return to them later instead of starting over. Open Chat history in the agent toolbar to browse previous conversations.

Chats are typically grouped by recency (for example, Today, Last week, 4 weeks ago, or Older) and may also surface chats that are related to what you’re currently looking at, so it’s easier to jump back into a relevant thread.

Linear Agent in comments

Outside of the dedicated chat experience, @Linear is also available directly in comments across issues, documents, and updates, as well as inline in project and initiative descriptions.

This is meant for moments when you’re already doing work and you want the agent to help in-place, using that context.

When you’re writing a comment, you can prompt Linear to help you produce something that’s ready to post — like a status update, a concise summary of progress, a list of action items, or a decision recap. This is useful when you want the output to live directly on the issue for team visibility.

Common comment prompts include:

Summarize the current state of this issue and call out blockers.

Rewrite this section to be more concise and action-oriented.

Turn these notes into a weekly update with progress, risks, and next steps.

Draft a summary of the overview for the next person picking this up.

Linear Agent for Slack

Linear Agent is available in both Linear and Slack, so you can collaborate on work in whichever tool your team is using.

In Slack, mention @Linear to take actions based on the context of your conversations.

Learn how to use Linear Agent for Slack

Skills

Overview

When a conversation with Linear agent yields a great result, save that workflow as a reusable skill.

Skills help you save time, get more consistent results when collaborating with Linear, and avoid rewriting the same instructions for repeated prompts or workflows.

Consider creating skills for workflows you use frequently, like:

You can create a skill by asking Linear directly to save a conversation as a skill for you. Once saved, skills can be invoked directly with a slash command in the agent input, and Linear Agent may also use them automatically when the context matches the skill’s intent. You can also create or manage skills in in Settings → Account → Agent personalization → Skills.

Best practices

Start in chat

Start by working with the Linear agent in a chat to solve a real problem.

Beyond achieving your immediate goal, this helps you figure out what instructions work, what context is needed, and what kind of result you want.

Save a skill once the result is good

Once you have an interaction that works well and you expect to use it again, ask Linear to save it as a skill. A good skill is repeatable and focused. It can have multiple steps, but it should do one job well. Avoid packing too much into a single skill.

Optimize the skill

After saving a skill, you can also keep improving it. You might ask Linear how to make it faster by excluding steps that aren’t needed every time.

Automations

Automations let you define how Linear should respond to issues entering triage using flexible, open-ended instructions.

Triage in Linear combines a few different systems that work together:

• Triage Rules are deterministic — they apply predefined actions when specific conditions are met
• Triage Intelligence analyzes issues and is best used for routing to the right team and individual, while also setting properties based on context
• Automations extend this by letting you define more complex, open-ended behaviors for how Linear should act on issues

Together, these allow issues to be processed, enriched, and acted on automatically without someone needing to prompt the agent each time.

Configure Automations in Team Settings → Triage → Agent behavior. From there, you can add a new automation, define the instruction Linear should follow, and optionally scope it to specific types of issues in your triage queue.

Guidance

Guidance is a set of instructions for how Linear operates within your workspace and across teams. Use guidance to make outputs consistent with conventions in your workspace. You might specify how you want issues titled, what fields should be included in bug reports, or how updates should be structured.

You can configure guidance for Linear at different scopes: workspace-wide, specific to a team or personal to you.

Workspace guidance applies to everyone in the workspace by default. Restrictions for who can edit workspace-level guidance is configurable by navigating into Settings → AI.

Personal guidance lets an individual tailor how Linear responds for them. Update your personal Linear agent guidance by navigating in Settings → Agent personalization.

If you’d like guidance to apply to a specific team only, you can reference that team and its unique instructions in your workspace-level guidance.

Preferences

When opening a new Linear tab, we launch a new chat with Linear agent.

If you'd prefer to change this behavior, you can navigate into your Preferences and update Default home view from Linear Agent to your desired location.