# Claude Cowork is closed source. Kortix is open source.

> No. Anthropic's support docs put Claude Cowork on paid plans (Pro, Max, Team, Enterprise) and on Anthropic's servers. Kortix is the open-source AI Operating System: an alternative you can read, fork, audit, and self-host.

URL: https://claudecoworkalternative.com/is-claude-cowork-open-source

## Claude Cowork and Kortix, side by side.

The Cowork column states what Anthropic documents on its own pages; the Kortix column quotes Kortix's docs.

| | Kortix (recommended) | Claude Cowork |
| --- | --- | --- |
| Open source | Yes | No |
| Who can read the code | Anyone. Read it, fork it, audit it | Anthropic only; no public Cowork source |
| Models | Any model provider with your own keys | Anthropic's Claude models |
| Where it runs | Kortix Cloud, your VPC, or your own on-prem network | Anthropic's servers, or your Amazon Bedrock, Google Cloud, or Microsoft Foundry account |
| Self-host | Yes. Self-host is free. | No |
| Where your configuration lives | In one git repo you own | In your Claude account's settings and saved plugins |

Cowork cells: Anthropic's product page and support docs, checked October 2026. Kortix cells: Kortix's own docs.

## What closed source means for a hosted agent

Claude Cowork is Anthropic's agent for everyday knowledge work: research, documents, spreadsheets, and scheduled tasks. Anthropic's [support documentation](https://support.claude.com/en/articles/13345190-getting-started-with-cowork) puts it on the paid Claude plans (Pro, Max, Team, Enterprise) and describes Cowork sessions running in an isolated environment on Anthropic's servers, with sessions and files saved to your Claude account. Anthropic's [product page](https://claude.com/product/cowork) adds one infrastructure option: enterprises can point the model at their own Amazon Bedrock, Google Cloud, or Microsoft Foundry account. What no Anthropic page offers is the software itself. No Cowork source code is published, so nobody outside Anthropic can read how the agent plans a task, runs its code, or handles your files. Configuration follows the same pattern: global instructions, connectors, and plugins are set inside Anthropic's interface and saved to your Claude account.

## What open source buys a team

Open source rewrites the answers a closed, hosted agent gives you. Your configuration stops being a settings screen: in Kortix, agents, skills, company memory, and connectors are files in one git repo you own, so you can version them, diff them, and take them with you. Model choice opens up, because Kortix works with any model provider with your own keys. Where the system runs is your call too. Run it on Kortix Cloud, in your VPC, or on your own on-prem network. Self-host is free. This site's [self-hosting guide](/self-hosting) walks through the setup. And the decision stays reversible, because the code is public: read it, fork it, audit it.

## Kortix is the open-source alternative

Kortix is the open-source AI Operating System: your agents, their skills, your company memory, and every connector in one git repo you own, with the agents working on real cloud computers. Any model. Self-host or cloud. One isolated sandbox per session. Session work reaches `main` through a change request. Merge is default-deny for agents. Kortix's public repo shows 20,000+ GitHub stars (October 2026). The [comparison page](/comparison) on this site lines the wider field up side by side, and the Kortix blog goes deeper: [read the Kortix blog](https://kortix.com/blog/kortix-vs-claude-cowork) for the head-to-head with Claude Cowork, or [the Kortix blog](https://kortix.com/blog/open-source-ai-agent-platform) for what an open-source agent platform looks like in practice.

## The configuration, as a file you own.

`kortix.yaml`

```yaml
kortix_version: 2
default_agent: kortix

project:
  name: my-project
  description: What this project is.

env:
  required: [DATABASE_URL]
  optional: [STRIPE_API_KEY, WEBHOOK_SLACK_SECRET]
```

Source: https://kortix.com/docs/project/manifest

The top of the manifest in Kortix's docs: project metadata and env var names, in the repo. Secret values stay out of the file.

## Frequently asked questions

### Is Claude Cowork open source?

No. Anthropic's support docs list Cowork on the paid Claude plans (Pro, Max, Team, Enterprise), with sessions running on Anthropic's servers. Anthropic publishes no Cowork source code, so there is nothing to download, inspect, or run on your own hardware.

### Is Kortix open source?

Yes. Kortix is the open-source AI Operating System: agents, skills, memory, and connectors in one repo you own. The public repo shows 20,000+ GitHub stars (October 2026).

### Can I read the Kortix source code?

Yes. The code is public: read it, fork it, audit it. It lives on [Kortix on GitHub](https://github.com/kortix-ai/suna).

### Can I self-host either?

Claude Cowork: no. Anthropic's docs put sessions on Anthropic's servers, or in an enterprise customer's own Amazon Bedrock, Google Cloud, or Microsoft Foundry account, with no self-hosting option. Kortix: yes, run it on Kortix Cloud, in your VPC, or on your own on-prem network; self-host is free, on one Docker Compose stack.

## Run your whole company from one repo you own.

Start with one job and grow from there.

[Try Kortix](https://kortix.com) · [Kortix on GitHub](https://github.com/kortix-ai/suna)

Open source · Any model, your keys · Self-host, VPC, or on-prem