Linear MCP
Linear MCP lets agents find, create, and update issues, projects, comments, and related planning objects in a Linear workspace.
Verdict
Common use cases
- Create tickets from Slack threads
- Triage bugs by priority and assignee
- Generate sprint progress summaries
- Update issue status in bulk
- Search backlog by label or milestone
Integration
- Vendor
- Linear MCP
- Category
- other
- Auth
- OAUTH2
- Composio slug
linear_mcp
Tools
Setup
Setup guide
- 11. Open your Switchy workspace settings and navigate to the Integrations tab. 2. Find Linear MCP in the directory and click Connect. 3. You'll be redirected to Linear's OAuth consent screen — sign in with your Linear account. 4. Grant read and write permissions for issues, projects, and comments (Linear will show the exact scope list). 5. After authorizing, you'll return to Switchy and see Linear MCP listed as active. 6. Open any Project and type '@Linear' followed by a test query like 'show me open bugs assigned to me' to confirm the connection works. 7. If the MCP responds with issue data, setup is complete and your team can start using it in any Project.
What teammates see: by default, memories from Linear MCP are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Triage Open Bugs
@Linear show me all open bugs in the current cycle, grouped by priority, and flag any that have been open longer than two weeks
Create Ticket from Notes
@Linear create a new issue titled 'Fix login timeout on mobile' with description from the notes above, label it as bug, and assign it to the on-call engineer
Sprint Progress Report
@Linear generate a progress report for the current sprint showing completed issues, work in progress, and anything blocked or overdue
Bulk Status Update
@Linear find all issues labeled 'needs-review' that have been in that state for more than three days and move them to 'in-progress'
Search Backlog by Theme
@Linear search the backlog for all issues related to authentication or login flows and show me the top five by priority
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Linear show me all high-priority issues assigned to the design team that are still in backlog
Found 7 high-priority issues in backlog for the design team. Top items: LIN-482 (Redesign onboarding flow), LIN-501 (Update component library colors), LIN-519 (Fix mobile nav accessibility). Each includes assignee, labels, and last update timestamp. This snapshot reflects your workspace state at query time—priorities and statuses may shift as the team triages.
Demonstrates filtered search across multiple dimensions (priority, team, status). Requires OAuth scope covering issue read access. Linear's API returns paginated results; this example assumes a manageable result set. If your backlog exceeds 50 issues, expect truncation or follow-up prompts.
@Linear create a new issue titled 'Add dark mode toggle to settings' in the Frontend project, assign it to me, and set priority to medium
Created issue LIN-847 in the Frontend project. Title: 'Add dark mode toggle to settings'. Assigned to you, priority set to medium, status defaulted to 'Backlog'. The issue is now visible in your Linear workspace and will appear in relevant project views. You can refine description, labels, or estimates directly in Linear or via follow-up prompts here.
Shows issue creation with multiple attributes in one prompt. Requires OAuth write scope for issues. Side-effect: the issue persists in Linear immediately—no undo via Switchy. If your workspace enforces required fields (e.g., estimate, labels), the MCP may return validation errors you'll need to address.
@Linear pull all completed issues from the last sprint and draft a release note summarizing new features and bug fixes
Retrieved 14 completed issues from Sprint 23 (Jan 6–19). New features: OAuth provider support (LIN-701), bulk export tool (LIN-688), keyboard shortcuts for search (LIN-702). Bug fixes: resolved 6 issues including pagination errors and stale cache warnings. Draft release note: 'This release introduces OAuth integration, a new bulk export workflow, and enhanced keyboard navigation. We also addressed critical pagination and caching bugs reported by users.' Adjust tone and detail as needed.
Combines Linear data retrieval with AI synthesis—useful for changelog generation or stakeholder updates. Accuracy depends on issue metadata quality (titles, labels, descriptions). The AI interprets 'feature' vs 'bug' from labels or issue type; verify categorization before publishing. Requires read scope across projects and cycles.
Use-case deep-dives
Linear MCP works when your standup lives in Slack or chat
A 5-person product team runs daily standups in Slack and needs to pull ticket status without opening Linear every time. Linear MCP is the right call here. The OAuth2 flow means each teammate authenticates once, then the team can query ticket states, assignees, and sprint progress directly in chat. The MCP shines when you're triaging 10-20 tickets per standup and need quick reads, not deep edits. If your team is larger than 12 people or you're managing multiple projects with complex filters, the lack of granular scopes in OAuth2 can become a security concern—everyone sees everything. For small teams with a single project board, this setup cuts standup time by 5-10 minutes per day.
When Linear MCP bridges support tools and engineering backlogs
A 3-person support team uses Intercom or Zendesk and needs to escalate bugs into Linear without switching apps. Linear MCP fits if your support reps are already working in a shared AI workspace and can trigger ticket creation with context from the support thread. The OAuth2 auth means each rep authenticates their own Linear account, so tickets get attributed correctly. This scenario works best when you're escalating 5-15 tickets per week—enough to justify the integration, not so many that you need a dedicated sync tool like Zapier. If your support volume is over 50 escalations per week or you need two-way sync (Linear updates flowing back to support tickets), the MCP's read-heavy design won't scale. For low-volume, high-context escalations, it saves 10 minutes per ticket.
Linear MCP is overkill for one-time sprint reporting
A 6-person remote engineering team wants to pull completed ticket counts and cycle time metrics for a bi-weekly retro. Linear MCP is not the right tool here. The OAuth2 setup requires each teammate to authenticate, which is friction for a workflow that runs twice a month. Linear's native CSV export or a read-only API key in a script is faster and doesn't require persistent MCP access. The MCP makes sense if your retros happen daily or if you're pulling live data during the meeting to answer ad-hoc questions. For periodic reporting, the setup cost outweighs the benefit. If your retros are more frequent than weekly and you need real-time ticket lookups, then the MCP starts to pay off.
Frequently asked
What does the Linear MCP do in Switchy?
It connects your Linear workspace to Switchy so AI agents can read issues, update statuses, create tasks, and query project data without leaving the conversation. You authenticate once via OAuth, then any team member can ask the AI to pull sprint backlogs, check issue dependencies, or draft new tickets based on chat context.
Do I need admin access to connect Linear?
You need workspace member permissions at minimum, but OAuth scope requests depend on what the MCP asks for during setup. If it requests write access to issues or projects, Linear may prompt you to confirm those permissions. Check your Linear workspace settings if the OAuth flow fails — some orgs restrict third-party app installs to admins.
Can it create issues in specific Linear teams or projects?
Yes, assuming the OAuth scopes include write permissions. You can tell the AI which team or project to target when creating an issue. The MCP reads your workspace structure during auth, so it knows your team names and project IDs. If you see permission errors, the connected account may lack access to that specific team.
Why use this instead of Linear's API directly?
The MCP wraps Linear's GraphQL API so you don't write queries by hand or manage tokens per user. In Switchy, the AI decides which Linear operations to call based on natural language — no code, no Postman collections. If you already have custom Linear scripts, keep using them; this is for teams who want conversational access without engineering overhead.
Does connecting Linear count against my Switchy plan limits?
MCP connections don't count as seats, but AI requests that call Linear tools consume your workspace's monthly message quota. Heavy usage — like bulk-creating issues or querying large backlogs — burns through messages faster. One team member connects Linear via OAuth; everyone in the workspace can then use it in shared chats.