For founders building with AI agents
CmdBrief: Stop juggling agents.Brief one lead instead.
CmdBrief is a Mac workspace where one lead agent plans the work, starts a crew with one job each, and brings you only the questions that are yours.
€29.99/month · 14-day money-back guarantee · Apple silicon · macOS 26+
- Claude Code
- Codex
- Grok CLI
- Kimi CLI
- OpenCode
- Cursor CLI
Why a lead
Six agents in six tabs is not a team.
Without a lead you are the switchboard. Every question, every approval and every “which file?” lands on you, one tab at a time.
Without a lead
You in the middle
With CmdBrief
A lead in the middle
Lead and crew
Your whole crew on one screen.
CmdBrief is a native Mac workspace for Claude Code, Codex, Grok CLI, Kimi CLI, OpenCode, and Cursor CLI. The lead sits on stage under the project’s phases, its crew is grouped by job beside it, and every decision reaches you through the lead. Switch to Grid to talk to any agent yourself.
InteractiveThis window works. Click anything, or let the tour show you around.
On it: five phases, six agents, each with one job. I’ll bring you only what needs you.
Update: the paywall is halfway. Luna has the view, Jude waits on Iris for the shared Snapshot type, and Seth’s tests are 8 of 16 green.
Iris sent three hero options. I need your pick before the paywall goes on.
One Tuesday with a crew
What a day with a crew looks like.
You spend the day on the parts only you can do. The crew takes the rest and tells you, through the lead, when it needs you.
You brief the lead.
One message to Remy, in plain words. No board to fill in, no twelve prompts to write.
The chat box
One box runs the whole crew.
The lead’s chat box and every agent’s work the same way. Same keys, same habits, one agent or six.
To the lead, or anyone
Pick who hears the next message: the lead, or any one of its crew, without leaving Stage.
/ commands and $ skills
Each CLI’s own commands and skills, listed as you type. Nothing new to learn.
@ files by name
Mention a file and the agent gets its path. No hunting, no pasting.
Screenshots as input
Paste, drop or attach up to four images. They reach the agent as real image input.
Queue, interrupt or stop
Type mid-turn and it waits its turn. When it can’t wait, interrupt and send.
Approvals where they’re asked
Allow once, for the session, or deny and say what to do instead. Or set the pane to auto-approve.
Not only code
A crew for the whole founder day.
You build, market, plan and watch the money. The crew takes a share of each, and every workspace keeps its own lead.
- Build
Ship the code
Builders write it, a Tester runs it on the simulator, a Reviewer reads every change before it lands.
- CREW
- Builder, Tester, Reviewer
- YOU
- approve what leaves the machine
- Market
Launch what you shipped
A Designer drafts screens and launch images in its own workspace, and another agent drafts the post.
- CREW
- Designer, launch copy
- YOU
- pick and publish
- Product
Keep the roadmap honest
Phases and exit gates live in your repo. The lead keeps them current as the crew checks gates off.
- CREW
- the lead, Review plan
- YOU
- own the gates marked you
- Money
Know what the crew costs
Per-agent usage as each CLI prices it, and a notice to the lead before a crew runs your allowance dry.
- CREW
- usage per agent
- YOU
- decide how hard it runs
The film
From one brief to what shipped.
Under a minute of the real app: one message to the lead, a crew of five on it, the one decision that comes back to you, answers as charts and plans, then where every token went and what shipped. Music only, so it plays anywhere.
Local-first
Your code never becomes our product.
CmdBrief is a local Rust app, not a cloud IDE, and it ships no usage analytics.
Read the privacy policy- Repositories, crew messages and run plans stay on your Mac. Agents get only the context you send.
- Secrets are referenced as env vars. Keys and tokens live in the macOS Keychain.
- It contacts cmdbrief.com to check your subscription. Crash reports go to Sentry (EU): the error, stack trace, app and macOS versions and Mac model, with your home path removed. CmdBrief adds no terminal content, prompts, responses, or files.
- Agent CLIs and the optional voice assistant reach their providers only when you use them, under your own accounts.
One plan
Every agent, every crew, one window.
€29.99/month
Billed monthly. Or pay yearly (€299.90) and save 17%.
Not for you? Email [email protected] within 14 days of your first charge and you get a full refund. One email, no questions.
- Lead and crew runs, up to 50 agents per lead
- Unlimited product workspaces
- Claude Code, Codex, Grok CLI, Kimi CLI, OpenCode, and Cursor CLI supervision
- Task-to-agent launch and session resume
- Voice agent support with your own keys
- Usage windows and limit awareness
- Priority support
Apple silicon · macOS 26+ · v0.5.35. Agent provider plans are separate and billed by each provider. Pricing details
Questions founders ask first.
Lead and crew
Any Claude Code or Codex agent in a Standard pane can be made the lead. It turns your goal into milestones with owners, starts crew agents with a job (Builder, Designer, Tester, Reviewer or Shipper), and gathers their reports. Crew members bring questions and design picks to the lead, and the lead asks you; only a member’s own permission prompt comes to you directly. Whatever waits for you is in Needs you, and you can still talk to any agent directly from the Grid view.
See lead and crew →Grok CLI, Kimi CLI and Cursor CLI run in terminal panes beside the crew with the same status, notifications and session resume. Each agent runs through its own CLI on your plan with that provider.
See lead and crew →All of them get the full supervision layer: live working, needs-action, completed, and failed states, notifications, and session resume. Usage meters cover Claude Code, Codex, Grok CLI, OpenCode Go plans, and Cursor CLI. Any other CLI can run in a regular terminal pane, but it does not report agent status.
See usage windows →
The basics
Every product gets a persistent home for its agents, tasks, terminals, files and browser context, with live agent status, usage meters and session resume.
See every feature →It is most useful when switching between products means switching repositories, agent sessions, tasks, browser tools, and operating context.
See use cases →Native pieces such as the menu-bar system monitor and notifications are part of why it is Mac-first.
We verify the email, then you can connect your workspace and start using CmdBrief on your products. One plan, monthly or annual, with a 14-day money-back guarantee on your first charge.
See pricing →
Voice
Voice works with a Standard pane the way it works with terminal agents: it reads the conversation, sends it a message, stops a turn, and answers its questions and approvals. Voice providers require your API key and bill usage separately.
See voice →It answers questions about agent status, workspaces, usage limits, terminal output, and project files using session-scoped content access, automatic or manual. It can start or resume supported terminal agents, work Standard panes, and send dictated instructions, translated into technical English before they reach an agent. With agent control enabled, voice actions can run immediately. Turn on “Confirm actions before they run” in Voice settings to review proposed actions locally. Opening a remote website still requires confirmation.
See voice →You can use a language supported by the selected provider and configure the assistant’s reply language. Language availability and accuracy vary by provider; each provider requires your API key and bills usage separately.
Privacy and tools
Agents connect to their providers under your accounts, and account, billing, and optional voice features can use network services. Session browsing uses local metadata plus bounded local prompt and response previews.
Read the privacy policy →You can coordinate launch checks, marketing deliverables, or traffic-driven implementation in one workspace, without pretending CmdBrief replaces your deployment, publishing, analytics, or search tools.
See daily goals →Custom connectors work too. Keep other analytics, search, CMS, or campaign tools in a browser pane and turn a finding into an agent task. A connector does not publish content.
See daily goals and connectors →
Still unsure?Write to [email protected] or use the contact form.Contact CmdBrief →
Practical guides
Choose your setup. Run the next task.
Compare the options, isolate parallel work, and keep the right session within reach.
- Claude Code GUI for Mac: the options compared (2026)The official Claude Code desktop app, Conductor, Nimbalyst, Superset, Sculptor, Claude Squad, cmux, and CmdBrief — what each one is for, what it lacks, and how to choose, verified against each vendor’s own docs.Read guide
- How to manage multiple Claude Code sessions on a Mac (2026)Six ways to run several Claude Code sessions at once — terminal tabs, --worktree, the built-in claude agents view, the desktop app, tmux tools, and CmdBrief — with the commands, a comparison table, and the failure each one is prone to.Read guide
- Run Claude Code and Codex in parallel with Git worktreesCreate separate worktrees, launch Claude Code and Codex, divide ownership, and review the combined result. Includes commands and a reusable task brief.Read guide
Understand Claude Code usage limits · Claude Code vs Codex · Browse every guide
Compare CmdBrief with other workspaces · CmdBrief vs Warp · CmdBrief vs Ghostty · CmdBrief vs iTerm + tmux + Claude Code · CmdBrief vs Cursor-as-workspace · CmdBrief vs Claude Desktop · CmdBrief vs Conductor · CmdBrief vs opcode (Claudia) · CmdBrief vs Emdash · CmdBrief vs JetBrains Air
Brief your first crew today.
Download CmdBrief, make an agent the lead, and tell it what you want shipped.
€29.99/month · 14-day money-back guarantee · Apple silicon · macOS 26+