Claude Code in NeuroSquadClaude Code,with a canvas around it
Run as many Claude Code sessions as you need, each in its own card with a real terminal. NeuroSquad knows when each one is working, waiting on you or finished, hands it browsers, terminals and notes by arrow, and keeps its sessions, limits and costs in view.
- The real
claudeCLI, your own login - Nothing written to
~/.claude - Free, for Windows 10 and 11
It knows when Claude is done — and when it’s asking
A quiet terminal can mean “finished” or “waiting for your permission”. NeuroSquad doesn’t guess: every Claude Code card reports its own turns through three of Claude Code’s hooks.
UserPromptSubmithookYou press Enter. The card’s icon spins, the Squad status card counts it as working, and the toast from its last question is taken down.
NotificationhookClaude stops to ask. The card pulses, a sound plays once, and the toast leads with what it asks. It stays until you answer.
StophookThe turn is over — not just quiet. The next queued prompt goes out, the journal gets the answer, and a toast says it’s done.
Click, and you’re there
A toast click switches to the workspace, flies the camera to that card and flashes it — even among dozens of agents.
One status everywhere
The card, the sidebar, the Squad status card, custom cards and the phone all read the same status.
Your hooks stay yours
The hooks come in a --settings file merged over your own settings, so hooks you set up still run.
Three things you’ll do every day
Replicas of the app, played step by step. When Claude asks for permission, the answer is yours.
Claude Code fixes a failing test, stops to ask before deleting a folder, and finishes — the journal writes itself.
Everything Claude Code can do in NeuroSquad
Claude Code has the deepest support of the nineteen agent CLIs. Here is all of it, with a docs page for each.
- 01
Knows its state
Claude Code’s own hooks report every turn, so nothing is guessed.
Finished and waiting are different
Each card gets three of Claude Code’s lifecycle hooks:
Docs: Finished and waiting are differentUserPromptSubmitmarks it working,Notification— waiting on you,Stop— finished.A notification that says what it needs
The toast leads with Claude’s own words —
Docs: A notification that says what it needsClaude needs your permission to use Bash. One per agent; a waiting one stays until you answer; a click flies to the card.Every card is its own session
The card’s id is the session id:
Docs: Every card is its own session--session-idon the first launch,--resumeafter, so restarting the app continues the conversation. “Restart session” gives the card a clean one.The Squad status card
Every AI agent of the workspace on one card — waiting on you, working, finished, idle — with its model from Claude Code’s own session log and the live time of its turn.
Docs: The Squad status cardCustom cards can follow it
Cards built with the Card SDK see the same statuses and turn events, and a card’s own tools reach a connected Claude Code over MCP.
Docs: Custom cards can follow it
- 02
Works through the canvas
An arrow from the card is a set of tools. Draw it, and Claude Code can use what is at the other end.
Arrows are tools
An arrow to a browser, terminal, note or another agent gives Claude Code
Docs: Arrows are toolsbrowser_*,terminal_*,note_*oragent_*over NeuroSquad’s MCP server — no restart. Each call lights up its arrow.It can add cards itself
Docs: It can add cards itselfcanvas_spawn_cardlets it open a terminal, a browser, a note or a helper agent next to itself, already joined by an arrow.Canvas mode
Claude Code starts without its own
Docs: Canvas modeBash,WebSearchandWebFetch(--disallowedTools), so commands run in terminal cards you can watch and the web opens in browser cards.MCP servers from the registry
Install a server from the official MCP Registry once, then give it to an agent with an arrow. Claude Code asks before every call to it, and its secrets never leave NeuroSquad.
Docs: MCP servers from the registrySkills from skills.sh
An arrow to a skill card gives Claude Code
Docs: Skills from skills.shskill_loadwith each skill and when to use it — marked to load up front, so it actually reaches for them.
- 03
Built for long runs
Hours-long sessions run into full contexts and usage limits. Both are handled on the card.
Context meter and Compact
A small ring in the header shows how full the context is and turns amber from 80%. Compact sends
Docs: Context meter and Compact/compactfor you.Continues after a usage limit
When Claude Code hits its usage limit, the card counts down to the reset and sends “continue” then — a night job carries on by itself. Off per card if you prefer.
Docs: Continues after a usage limitPrompt queue
Line up the next messages: each goes out when the turn is finished — and with the hooks, “finished” is exact, never a pause mid-thought. Daily prompts too.
Docs: Prompt queueHand off to a fresh agent
When a session is full, start a new agent beside it — Claude Code or any other CLI — with a summary of the work typed in, ready for you to send.
Docs: Hand off to a fresh agentJournal
Each finished answer is added to a connected note: a running log of what the agent did, written without asking the agent to.
Docs: Journal
- 04
Works in a team
Several Claude Code agents — or a mix of CLIs — on one project without stepping on each other.
Isolated worktrees
Give an agent its own git worktree, so parallel agents never edit the same files. A setup command and files to copy, like
Docs: Isolated worktrees.env, prepare it before Claude Code starts.Add a reviewer
One click starts an agent on a different CLI in the same worktree, with an arrow from the author and a prompt to review without editing anything.
Docs: Add a reviewerDangerous mode, per card
For a run you trust, a card can start Claude Code with
Docs: Dangerous mode, per card--dangerously-skip-permissions. It’s a per-card switch, off by default.Roles
Reviewer, tester, researcher, architect, writer or your own: the role is typed in as a first prompt you can see and edit, and never re-sent behind your back.
Docs: RolesOne instructions file
You edit AGENTS.md; NeuroSquad mirrors it to CLAUDE.md and never overwrites a changed file without showing you the diff.
Docs: One instructions file
- 05
Cost, control and access
What it spends, what it may spend, which model it runs on, and how you reach it.
Exact usage and cost
Tokens and cost by card, workspace and model, read from Claude Code’s own transcripts, subagents included. Whole tokens, costs exact to the picodollar.
Docs: Exact usage and costBudget brake
A limit per workspace. Past it, each agent gets Ctrl+C — the session is kept — and automatic prompts stop. Raise the limit and it’s lifted; your own typing is never blocked.
Docs: Budget brakeOpenRouter models
Run a card on an OpenRouter model instead of your Anthropic login. The key goes only into that card’s environment, never into arguments or logs.
Docs: OpenRouter modelsDictation into the prompt
Hold a hotkey and speak. Speech is recognised on your machine and typed into the active agent’s prompt.
Docs: Dictation into the promptFrom your phone
Scan a QR code and the whole canvas opens in the phone’s browser: read Claude Code’s terminal, answer its question, send the next prompt.
Docs: From your phone
Who’s waiting, and what it cost
Two cards that answer the questions you ask most when several Claude Code sessions run at once.
The Squad status card
Every agent of the workspace with its status, model and turn time. The one waiting on you comes first, with what it asks. Tap a status to filter.
- coupon-fixClaude CodeOpus · 1MClaude needs your permission to use BashNeeds you2:34
- api-migrationClaude CodeOpus · 1MMove routes to the new clientWorking14:58
- db-indexesClaude CodeSonnetAdd the missing indexesWorking6:42
- docs-passClaude CodeSonnetUpdate the READMEFinished4 min. ago
Usage & cost
Read from Claude Code’s own logs on your disk and matched to the card by session id. The rows always add up to the total.
| Agent | Tokens | Cost | ||
|---|---|---|---|---|
| coupon-fix | 1,284,310 | $3.12 | ||
| api-migration | 842,905 | $1.87 | ||
| docs-pass | 216,774 | $0.41 | ||
| Total | 117 | 2,343,989 | $5.40 |
The real Claude Code, plus a few arguments
NeuroSquad starts the same Claude Code you run yourself, in a real terminal in the card. This is everything it adds:
Nothing written to ~/.claude
The hooks and the MCP config are per-card files in NeuroSquad’s own data folder, passed as arguments.
Your login, your plan
Claude Code runs with the subscription or API key it already uses. NeuroSquad has no account and no cloud.
Found where you installed it
The native installer’s claude.exe first, then npm’s. PATH is read fresh, so a CLI installed while the app is open is found by the next card.
Claude Code in NeuroSquad, answered
Give Claude Code a canvas
Free for Windows 10 and 11. Your code, your login and your costs stay on your machine.