How Rundock compares to Claude's own products
Most AI tools give you a chat window. You open it, ask a question, and close the tab. It's useful, but it forgets you every time.
Rundock takes a different shape. It runs on Claude, the same models you already use, but instead of one window it gives you a team. Named agents, each with a role, that remember your work and follow your rules.
So this isn't Rundock against Claude. Rundock runs on Claude. The real question is how you want to work with it: something you talk to, or something you run.
Which Claude is which
The Claude desktop app holds Claude, Cowork, and Claude Code at once, which is part of why they're easy to confuse. Here is what each one is, and where Rundock fits.
| Claude | Claude Cowork | Claude Code | Rundock | |
|---|---|---|---|---|
| Holds the context? | The conversation holds it. A new chat starts fresh. | Preferences persist. Each task starts from the brief you give it. | Config files hold it. Durable, but hidden by default. | Your vault is the shared, visible state. It persists. |
| Owns the work? | You, in one window. | You delegate a task, Claude returns a deliverable. | You can define agents, in files you rarely see. | Named agents each own a domain. You know who to go to. |
| What's it like to use? | A single chat. Simple and familiar. | A workspace in the Claude desktop app. | A command line, or a developer-facing app. | One place you navigate, with your team on an org chart. |
| Price checked July 2026 | Free to start. Pro from about $20/mo for more usage. | Needs a paid plan (Pro or Max). | Included in Pro or Max. Optional metered API beyond your limits. | Free and open source. Runs on your Claude Pro or Max plan, so no extra cost. |
| Effort over time? | No setup. But you re-brief from scratch each session. | Little setup. Each task still starts from your brief. | The setup is learning the terminal. Fast once you're over it. | More to set up once. Then every session builds on the last. |
| Best for | Quick questions and everyday help. | Handing off a defined task for a finished result. | Developers who want a capable agent in their workflow. | Running a team of agents you can see and manage, no code. |
Prices and capabilities change. Checked July 2026; verify current details at claude.ai/pricing.
Rundock is free and open source. If you already pay for Claude Pro or Max, you have everything you need to start.