Invite codes no longer gate your own accounts. Bring-your-own-account use is free for everyone; cloud Pro stays invite-only.Read the announcement
AnnouncementAugust 9, 2026

To everyone who's been following or using Mirasim:

Over the past few days, Mirasim ended up in front of a much bigger audience much earlier than we'd planned. We were still in private beta and, honestly, nowhere near ready for a wider launch.

Invite codes spread far beyond the small beta group we had planned for. In a very short window, we saw a flood of visits and test requests, along with some unusually aggressive, probing request patterns. Our capacity, reliability, and security systems weren't ready for that volume. Some of you ran into slow loading, failed requests, or Pro benefits that never showed up.

We're genuinely sorry for the rough experience. Whatever the source of the traffic, we should have been better prepared. That's on us, and our users shouldn't have to pay for it.

At the same time, you sent us a huge amount of honest, detailed feedback. A lot of issues that probably wouldn't have surfaced until a public launch showed up early. Thank you to everyone who took the time to try Mirasim, patiently share feedback, or report security issues responsibly.

Here's how access to Mirasim will work from here

Invite codes no longer limit connecting your own accounts to Mirasim — they only limit the cloud Pro service that Mirasim pays for.

Connecting and using your own accounts locally will be free for everyone. You'll be able to connect the Claude Code, Codex, and other model accounts or services you already use, then work across multiple models, Agents, and sessions in Mirasim. Mirasim won't charge for any of this. Third-party subscription or API fees will still follow each provider's own terms.

Cloud Pro access — where Mirasim covers the model usage costs to make things cheaper for users — will remain invite-only and roll out in batches as capacity allows. We'll keep sharing our progress and what opens next, so the rules stay as clear and transparent as possible.

A note on Pro benefits

  • The 1,000 beta users covered by our original promise will still receive one full year of Mirasim Pro. That promise has not changed.
  • Anyone who successfully registered during this wave after we had already passed the original 1,000-user limit will receive one month of Mirasim Pro. This is both to make up for the unstable experience and to thank you for trying Mirasim before it was fully ready.
  • If you've already registered or activated your access but your Pro benefits still aren't showing up, please don't try again. We have the records. We'll verify eligibility and apply the benefits in batches, then share the timing and start-date details.

How the rollout will work from here

Over the next week or two, we'll focus on adding capacity, fixing the entitlement system, tightening security, and working through your feedback one issue at a time. Mirasim remains open. Registration and local, bring-your-own-account access will stay available, while cloud Pro access will roll out in batches as capacity allows.

We're not trying to build yet another model aggregator. We want Mirasim to be an Agent workspace you can keep building in for the long run. Wherever the next leap in Agent capability comes from, you should be able to plug it into the same workflow, switch seamlessly, and use different Agents together. Once you've built something, you can put it in front of simulated users in a real environment and let evidence — not guesswork — tell you what to do next.

We want every leap in what Agents can do to become real leverage for everyone — so you spend less time chasing tools and more time building what you actually care about.

Models will change. You shouldn't have to start over.

Thank you for showing up before Mirasim was fully ready. From here, we'll put these promises into every fix and every update, and work to earn your trust by making Mirasim more stable and more reliable.

— The Mirasim team

Mirasim
Get helpGet Mirasim

Mirasim Documentation

Chapter 04

Parallel sessions

Several agents talking at once, and a way to switch between them that does not lose your place.

Running more than one agent

Every session is a pane, and panes run at the same time. Nothing about a second session slows the first: separate agent processes, separate transcripts, separate queues. What you are actually managing is not the agents' throughput — it is your own attention.

Two shapes cover most real parallel work:

  • Parallel slices — give each pane a different, non-overlapping part of the job (one on the API, one on the UI, one on tests). Merge them in order, smallest first.
  • Competing drafts — give two or three panes the same task with different models or different framings, keep the best result and discard the rest. On an ambiguous problem this is usually the fastest route to a good answer.

Switching between agents without losing your place

The sidebar lists your workspaces and their sessions; you can scope it to the current workspace or go global, filter it, pin the sessions you keep coming back to, and rename any of them (clear the title to get the automatic one back). Sessions are ranked by what needs you rather than by clock: the ones waiting on an answer first, then the ones still running, then everything else by recency. The numbered jump chords follow that same order, so ⌘1 is whatever most wants your attention.

Getting between sessions
The session sidebar with several workspaces open, the search entry point, and the new-session control.

A session rowOne row per conversation: its title, which agent is running it, when it was last active, and a status dot. — this control is not on the captured screen.

The buttons in this panel work — click them to look around.

  • ⌘KCommand palette: sessions, workbench commands, and full-text transcript matches in one list
  • ⌘1…through ⌘9 — open the Nth session in that attention order as a pane
  • ⌥1…through ⌥6 — focus pane N inside the current group
  • ⌘⌥←Cycle focus between panes (⌘⌥→ the other way)
  • ⌃⇥Cycle workspace groups — whole scenes, not panes (⌃⇧⇥ backwards)

Reading a running session

A transcript is not a chat log with the interesting parts hidden. Each turn shows the agent's thinking, its tool calls and their results, any sub-agents it delegated to, and the plan it is working through — collapsed by default, expandable when you want the detail.

What the transcript surfaces
Thinking / tool calls
The reasoning and every tool invocation with its result, expandable inline. Long output collapses with its character and line count.
Sub-agents
A delegated unit of work shown as its own block: the brief it was handed, the trail of what it did, the result it handed back.
Plan
The agent's todo list with live progress — how many tasks are done, in progress, or incomplete.
Live usage
A per-turn line: context used, cumulative tokens, output this turn. It is how you notice a session getting expensive before the bill does.
Notes and status
Give a session a status (Inbox, Active, Review, Done, Parked), record what it produced (Answer, Proposal, Implementation, Partial, Blocked, Conversation), and leave yourself a note.

Knowing which session needs you

With five panes open, the only question that matters is which one is waiting on you. Sessions carry an explicit state rather than leaving you to infer it from scrollback — and when a session goes quiet, Mirasim says why instead of pretending it is still working.

Running
A turn is in flight.
Needs you
The agent asked a question, requested a permission, or a decision is blocking it. Answer in place — no restart.
Queued
Your next prompts are lined up behind the current turn.
Stalled
Nothing has moved for a while. The reason is spelled out: quiet for N seconds, a named tool still running, awaiting the model, or the process merely alive.
Done / failed
The turn settled, or it ended without a reply or with an error — stated as such, not smoothed over.

Answering an agent mid-task

When an agent needs a permission or a choice, the question appears in the session, numbered if there are several. Pick one of its options, type a custom answer, or skip it. Answering does not restart the turn — the agent continues from where it paused.

Context, compaction, and long sessions

Long sessions run into the context window. Mirasim shows how much of it a turn used, and the conversation can be compacted — summarized down to free room — either on demand from the composer's command list or automatically when the window fills. Either way the transcript records that it happened and how much was freed, so a suddenly forgetful agent is never a mystery.

Ending and cleaning up sessions

Sessions are records, not scratch buffers — so deleting one asks what you actually mean.

  • Mirasim record only — keeps the agent CLI's own transcript, so you can still resume the conversation outside Mirasim.
  • Also delete the agent transcript — removes the agent's own jsonl / rollout files too. The conversation is fully gone.
  • Also remove the worktree — for a session that ran in one. Git refuses to remove a worktree with uncommitted changes, which is the correct behavior, not an error to route around.

A running session has to be stopped before it can be deleted. If you want to pick a conversation up in the agent's own CLI instead, copy its native resume command from the session menu.

Something wrong or missing on this page? Tell us