Tools· Operations & Project Management

    Linear

    Issue tracking and planning tool for software teams that organizes work into cycles, initiatives and roadmaps, automatically routes customer feedback into prioritized issues, and connects to coding tools through an MCP server and its own AI agents.

    executableplanningstructuringFreemium

    Description

    Strengths

    Consistent work hierarchy
    Initiatives, projects, cycles, and issues, together with teams and sub-teams, form a structure that connects the roadmap level and individual tasks in the same interface.
    Tight code connection
    Structural diffs for pull requests can be viewed, commented on, and linked to the corresponding issue directly in Linear, without switching between tracker and repository.
    Automated processing of customer feedback
    Conversations and feedback, for example from Gong recordings, are turned into titled issues backed by transcript excerpts and land in the relevant team's triage view.
    Broad integration coverage
    The integration directory connects tools including GitHub, GitLab, Slack, Figma, Notion, Salesforce, Sentry, Zendesk, and Zapier directly to issues and projects.
    Peek preview for quick review
    A keyboard shortcut opens a detail preview of issues or projects directly from any list or board view, without leaving the page.

    Assessment

    AI features

    • Linear Agent for delegated issues Teams can build and deploy their own AI agents that work alongside them on shared tasks or handle entire issues independently from start to finish.
    • Coding sessions with controllable agents Diffs can be reviewed on mobile and individual lines commented on to steer the ongoing coding session, together with iterating with Linear Agent, signed commits, and the option for workspace admins to require a signing key.
    • Loops for recurring work A job described in plain language runs on a schedule or in response to an event and uses context from the workspace and connected tools, for example to review incoming issues, turn unstructured notes into follow-up work, or keep launch plans up to date.
    • Agent-assisted text editing Changes made by Linear Agent to documents and project descriptions are highlighted separately and can be reverted to an earlier state through version history.
    • MCP server for external AI assistants A Model Context Protocol server exposes Linear data to compatible models and agents, either with write access or through a dedicated read-only endpoint or a restricted OAuth scope.

    Suitable for

    • Software teams that want to manage roadmap, projects, and individual issues in one consistent structure instead of separate tools
    • Teams that want to tightly couple pull request reviews and coding sessions to their issues while increasingly delegating tasks to AI agents
    • Organizations that want to turn customer feedback from conversations into prioritized, routed issues automatically
    • Less suitable for Less suitable for teams outside software development that mainly need a simple task board without a code connection or agent features

    Limitations and notes

    • Key features tied to Business/Enterprise SLAs, private sub-teams, and the free-of-charge Guided Reviews require a workspace on the Business or Enterprise tier, while the Gong integration is limited to Enterprise workspaces.
    • Loops run on AI Credits The Loops automation feature consumes usage-based AI credits; at launch a promotional credit of 20 USD per seat was granted, which according to the announcement expires on August 20, 2026, as of August 2026.
    • Watch the API rate limits The GraphQL API caps unauthenticated requests at 600 per hour and limits individual queries to a complexity of 10,000 points; polling the API for updates is explicitly discouraged; webhooks should be used instead.
    • Gong integration only for external, longer calls Internal or private calls as well as recordings under ten minutes are automatically skipped; only customer-facing conversations are processed.
    • MCP access can deliberately be limited to reading For AI assistants accessing Linear data, a dedicated read-only endpoint and a restricted OAuth scope are available alongside the full read-write endpoint, useful as a precaution for automated access.

    Quick start

    1. Create a Linear workspace and set up teams for your own work areas.
    2. Create a first project or initiative and add the corresponding issues, or import them from an existing tracker.
    3. Enable cycles to pace work, and set up board or timeline view to match your own way of working.
    4. Connect relevant integrations, such as GitHub or GitLab for pull request reviews and Slack for notifications.
    5. If needed, connect the MCP server to an AI assistant, starting with the read-only endpoint.

    Tips

    • Use peek with the space bar to quickly check issues and projects instead of fully opening them for every glance.
    • Keep SLA rules lean and review the built-in default rules first before adding custom ones.
    • When giving AI assistants MCP access, start with the read-only endpoint or the restricted read scope and only widen to write access deliberately.
    • Use webhooks instead of repeated API polling to track changes automatically, to stay under the rate limit.

    Access

    Last reviewed: · Pricing, plans and features are a snapshot in time. Check the provider's own page before deciding.

    In the workshop this becomes your method.

    Whoever sees this process run once wants the agent behind it next. We build that in the workshop From Process to Agent.

    View workshops

    Related resources

    Browse all resources

    Conversation, not pitch

    Understand first, then decide. We take time for an initial conversation, without sales pressure, without obligation.

    Schedule a call