What happens at issue 251 on Linear's free plan

You notice it as an error, not a warning. Someone goes to file a bug, hits create, and Linear tells them the workspace is at its issue cap. Whatever warning existed, nobody read it. The create button just stops working.

Linear's free plan allows 250 non-archived issues. Members are unlimited, which is genuinely generous, and you get 2 teams and 10 MB file uploads. But the 250 is a hard stop: until someone archives old issues or the workspace upgrades, nobody creates anything new. For a team shipping at any real pace, that is a few weeks to a few months of work.

What Linear's 250 issue limit actually counts

The word "non-archived" is doing all the work in that sentence. A bug you closed six months ago still counts against the 250 until someone archives it. Finished work occupies the budget.

So the free plan comes with a chore attached. Someone has to sweep completed issues into the archive to keep room for new ones. Nothing is lost when they do, archived issues stay searchable, but it is a task that exists only because of the cap, and it is exactly the kind of task a team forgets about until creation breaks on a Thursday afternoon.

The three doors at issue 251

Archive. Free, and fine if your closed-issue pile is genuinely dead. It buys you weeks, not years, and you will be standing here again.

Pay. Linear Basic is $10 per user per month. This is where the shape of your team starts to matter, because Linear charges per member, not per person who writes code. The designer who files two issues a month, the founder who reads the board on Mondays, the support lead who watches bug tickets: all $10.

Move. Most teams do not, because they assume migration means losing history. That has not been true for a while.

The math for a team of eight

Take six people who build and two who watch. A support lead who needs to see bug status and a founder who reads the roadmap. Neither will ever drag a card.

	Who pays	Per month	Per year
Linear Basic	all 8 at $10	$80	$960
Taskfolk Pro	6 builders at $3, viewers free	$18	$216

The interesting number is not the $744 gap. It is the two watchers: $240 a year on Linear, nothing on Taskfolk. That is the part that scales badly. Every stakeholder you add for visibility is another $120 a year for read access, so the instinct to open the board up to the whole company arrives with a bill.

Our free plan has no issue cap and no member cap, but it has its own ceilings and you should hear them from us rather than find them: 5 projects, 128 MB of workspace storage, 1 MB per file, 25 AI credits a month, and imports of up to 1,000 rows per run. If your team attaches large files to everything, 128 MB disappears quickly and you will end up on a paid plan for storage instead of for issues. What you will not hit is a wall on the thing a tracker exists to hold. Full numbers are on the pricing page.

Moving without losing the history

The importer takes a Linear API key. Paste it, pick what to bring, and issues, statuses, assignees, labels, and comments come across. A design agency that switched earlier this year moved 172 issues in about two seconds, which is roughly how long it takes to read this sentence.

What lands where:

From Linear	In Taskfolk
Issues	Issues with title and description
Workflow states	Board status columns
Assignees	Assignees
Labels	Labels
Comments	Comments on the issue

Two practical notes. The free plan imports 1,000 rows per run, which is four times the entire Linear free cap, so a team arriving from a full free workspace fits comfortably. And the boring migration process is still the right one: bring one project over, run it for two weeks next to Linear, then move the rest or decide not to. Nobody has to declare a switch on day one.

AI agents as named members

If part of why your issue count climbed is that agents are now filing and closing work, a per-issue ceiling is an awkward place to meet that trend.

On Taskfolk an AI agent is a member. Claude Code, Codex, and Cursor connect with their own API key and get their own profile, so work assigned to an agent is assigned to a name, its runs show as sessions on the ticket, and every change it makes is attributed to it in the activity log. Agent members are never billed as seats.

That is a structural difference rather than a feature-list one. On a strict per-seat tool, giving an agent an identity means paying for another user, which quietly pushes teams toward running agents under a shared human account where nobody can tell afterwards who did what.

Linear is not the villain here

Linear is fast and unusually well built. Its keyboard-first triage is the best in the category, the cycle model is coherent, and Insights gives engineering leaders analytics we do not match today. None of that changes because of a cap.

And the cap is a business decision, not a flaw. Free plans exist to become paid plans, and Linear picked issue count as the lever. It is at least a legible lever, arguably more honest than a free tier that works fine until you need one specific feature. If Linear's speed and its cycle workflow carry how your team works, pay the $10 and stop reading comparison posts. That is a real answer, and for some teams it is the right one.

The question worth asking is narrower than "which tool is better." It is: are you about to start paying per seat for people who only read? If seven of your eight are builders, the difference is small enough to ignore. If half the workspace is watching, the answer changes, and it keeps changing every time you add someone.

If you want the side-by-side rather than the argument, Taskfolk vs Linear lays out both columns including what Linear still does better. If Linear is not the only tool you are weighing, Linear alternatives covers where teams actually land. And if you just want to see whether your issues survive the trip, the importer runs on the free plan, so you can find out before you decide anything.