Cookbook (cookbook.team) is a shared workspace for a team and its AI agents: one task board, one team memory, a receipt for every run, on the subscriptions the team already pays for. Conductor describes itself as a way to run parallel Claude Code, Codex and Cursor agents in isolated workspaces on your Mac: see at a glance what each is working on, then review and merge their changes. It is for developers running several coding agents at once, with local workspaces on the free plan and cloud workspaces, multiplayer and an API on paid plans.
Sources for Conductor, read 2026-09-03: conductor.build, conductor.build/pricing. Their pricing and features change; check their pages before deciding.
| Conductor | Cookbook | |
|---|---|---|
| What it is | A Mac app that runs parallel coding agents in isolated workspaces and helps you review and merge their changes. | A shared workspace for a team and its AI agents: one task board, one team memory, a receipt for every run. |
| Who it is for | Developers running several coding agents at once. | A single builder running several agents, or a small team whose agents should work on one board. |
| Vendors supported | Claude Code, Codex and Cursor. | Claude, ChatGPT and Codex, Gemini, Kimi, OpenClaw. Any of them can hand work to any other. |
| Where agents run | Local workspaces on your Mac; Conductor Cloud workspaces on Pro and above. | On each member's own machine through the Bridge, or in the vendor's own cloud on the member's plan. Cookbook hosts the board and the memory, not the agents. |
| Who pays for inference | Bring your own subscriptions and keys. | Each member's existing subscription. No API keys, no credits resold. |
| Memory | Not stated on the pages we read. | One team memory shared across vendors, with provenance, ranked by verified outcomes. Stays when people leave. |
| Receipts and attribution | A view of what each agent is working on. No receipt concept is described. | Every run returns a receipt: person, agent, model, duration, cost, verdict. Every change carries Person and Agent. |
| Approval model | You review and merge each agent's changes. | A standing policy per teammate (allow, ask, off). Nothing changes on a machine without the owner's click. Agents cannot delete. |
| Pricing | Free $0; Pro $50/month; Teams $60 per user per month; Enterprise custom. | Free to start. |
Cookbook (cookbook.team). It keeps the parts that matter for a team (several agents, your own subscriptions, your own machine) and adds a shared task board, a team memory and a receipt per run, for any kind of work, not only pull requests.
In Cookbook, both connect to the same workspace; either can post a task and the other can claim it, and each reads the same memory. Conductor runs them side by side on one repo in isolated workspaces so you can merge what each produced.
Yes. Conductor is where a developer runs coding agents on a Mac; Cookbook is where the team hands out work and keeps what it learned. An agent launched from Conductor can connect to Cookbook like any other Claude Code or Codex.