Switch from Linear agents to Task Machine
Linear agents keep delegated work, Loops and cloud coding inside Linear. Task Machine gives people and agents Tasks with plans, review gates and Inbox decisions across your business.
Prefer the side-by-side comparison?Why people switch
-
App agents cannot lead
Linear's app-agent API does not let agents become project members or project leads, even with the assignment scope.
-
Coding needs GitHub repositories
Linear's native coding environments require repositories available through the workspace's GitHub integration. Each environment contains one repository.
-
Unused credits expire
Purchased AI credits expire 12 months after purchase. Added funds are non-refundable and non-transferable, and failed runs, retries and partial completions consume resources you pay for.
What maps to what
Every answer links to its source and shows when we last checked it. The oldest check on this page was November 15, 2026.
| Area | In Linear agents | In Task Machine |
|---|---|---|
| Tasks |
In Linear agents
An app agent becomes an issue's delegate rather than its assignee, so the human keeps ownership. Sessions track the agent's work and visible activity. |
In Task Machine
A person or an agent owns each Task. Its plan, discussion, decisions, runs and history stay on the same record, such as a bug fix or a weekly report. |
| Recurring work |
In Linear agents
Loops run on schedules or supported events from Linear, Slack, GitHub and external webhooks. Each loop has run history showing its actions. |
In Task Machine
Schedule a Task or Workflow in your timezone. A recurring job keeps one root Task, with a separate status and history for each occurrence. |
| Approval |
In Linear agents
Choose which connector tools a loop can use. Require approval for calls or automatically approve selected tools or an entire connector. |
In Task Machine
Set each agent's actions to Ask first or Automatic. Decisions arrive in your Inbox with context, and a Human approval step can hold a report before it is sent. |
| Guidance |
In Linear agents
Save a successful conversation as a personal or team-shared skill. Linear Agent can invoke skills automatically when the context matches their intent. |
In Task Machine
Agents keep reusable lessons from corrections in Memory, such as an accepted example or a mistake to avoid. You can inspect, correct and restore the notes. |
| Workflows |
In Linear agents
Configure a loop with instructions, a trigger, connectors and access settings. Workspace templates reuse eligible configurations, and published versions can be restored. |
In Task Machine
A Workflow defines steps and branches, including human questions and approvals. Add verifier checks to send a failed result back with feedback before the next step. |
| Code review |
In Linear agents
The native coding agent writes and tests changes, creates draft pull requests and addresses review comments. GitHub permissions and review requirements determine what ships. |
In Task Machine
Agents address review feedback and update branches under repository rules. Task Machine records approval of the exact commit before an agent may merge. |
| Execution |
In Linear agents
Native coding runs in cloud environments with repository-specific tools, setup scripts and instructions. Your team does not need a local development environment. |
In Task Machine
Use Managed Cloud with nothing to install, or connect a Local Worker for files, tools and accounts on a computer you control. Choose where each agent runs. |
| Pricing |
In Linear agents
Linear Agent is included on Free. Loops and cloud coding need prepaid AI credits. Coding bills model tokens at provider rates without markup, plus sandbox runtime, unless your own provider bills model usage. |
In Task Machine
Usage based, from $29 a month before tax. The whole payment becomes usage that your workspace shares, with no base fee. |
A fact we cannot find or cannot check is left out, and prices can change after the date shown. Task Machine publishes this comparison and is an interested party. Linear agents is named only to identify it, and there is no relationship between the two companies. Has something changed? Tell us.
You may stay with Linear agents if
You may stay with Linear agents if your team wants agent work to stay beside its existing issues and review discussions. The native coding agent lets teammates follow progress, steer changes and review diffs in Linear. Loops already handle scheduled summaries, incoming bugs and external events, so recurring work alone is not a reason to move.
What you give up
Task Machine's docs do not describe importing Linear Loops, skills or agent sessions. Rebuild the instructions and permissions, reconnect accounts and test the outputs. You also move the selected job's daily review away from Linear's issue and coding-session context. Keep Linear for planning if that context still serves your team.
There is no trial. Your first payment funds usage, and ordinary unsuccessful attempts still consume it. Check Billing before moving an expensive loop.
How the switch works
- Keep an archive before changing the automation. Linear documents issue CSV exports and copying issues or documents as Markdown. CSV exports omit attachment files, so save needed attachments separately. Copy each selected loop's instructions, trigger, tool permissions and reusable skill guidance into your migration notes.
- Create your workspace, connect the required repository and service accounts, and put reference documents in the Library. Choose Managed Cloud or a Local Worker for the job.
- Rebuild one job as a Task, Playbook or Workflow. Turn a weekly summary loop into a draft report, a verifier check and a Human approval before sending.
- Set autonomy to Ask first for unfamiliar actions. Compare a test result with the old process and check its Run history before allowing live writes.
- Set the schedule and timezone or reconnect the event source. Disable the original loop before enabling the replacement's live actions. Retain the original configuration until the replacement works reliably.
Common questions
Do I need to leave Linear's issue tracker?
You can move one agent job while keeping planning in Linear. This guide concerns Linear Agent, Loops and native cloud coding, not replacing every issue-tracking process. Use Connectors for the services the new job needs.
What happens to my data and skills?
Keep Linear exports as an archive and save attachments separately. Do not assume an issue archive recreates an agent configuration. Task Machine's docs do not describe importing Loops or skills, so rebuild the process and retain source material in the Library.
How do the costs compare?
Linear separates usage-based Loops and coding from its workspace plan, with model and sandbox charges for coding. Task Machine's pricing starts at $29 a month before tax, with the whole payment available as shared workspace usage and no base fee or trial. Compare the same job's actual usage rather than equating an issue with a Task. You keep 100% of your revenue, Task Machine takes no cut and never custodies your accounts.
Can I keep using my own provider account?
Linear's coding guide supports provider API billing, while sandbox runtime still uses Linear credits. Task Machine's Local Worker uses tamacode and can use the model subscriptions you sign in to on your computer. It is not a host for Linear's coding sessions.