Most AI tools in the workplace start and end with a chat box. You ask a question, get an answer, and then move the result to another application to complete your task.
Cloudflare wants something different.
The company released Cloudflare OS, an open source, AI-powered workspace platform designed for creating documents, presentations, spreadsheets, workflows, and custom applications, while connecting to the systems a business already uses.
Despite the name, it's not an operating system like Windows, macOS, or a Linux distribution. Cloudflare OS is closer to an AI-powered office suite, though even that description doesn't fully capture what the company is attempting.
Instead of giving employees a fixed collection of apps, Cloudflare OS lets them ask an agent to build tools for specific tasks. An employee could request a project dashboard, a reporting system, a collaborative whiteboard, a customer presentation, or an issue tracker without first finding and developing separate software.
Cloudflare calls these custom applications Gadgets. Each can include its own interface, server-side logic, API, and stored data. Users can share a working Gadget with colleagues or distribute a Blueprint that allows others to create their own separate copies.
The entire project is available through Cloudflare’s official GitHub repository as Cloudflare OS. Organizations can deploy it to a Cloudflare account, connect internal services, choose supported AI models, and customize the interface and instructions for how the company operates.
https://github.com/cloudflare/cloudflare-os
Cloudflare has already tested the basic idea on its own workforce. CIO Sam Rhea explained in a separate post on Cloudflare’s internal AI usage that employees wanted agents capable of interacting with corporate systems, but providing API keys would create serious security risks. That concern shaped much of Cloudflare’s OS.
Agents and created applications start with no access to anything. A user must grant permission to a specific resource, such as a GitHub repository, document, dataset, or internal service. Credentials remain outside the scope of the agent.
Cloudflare manages these connections through services it calls Gatekeepers. A Gatekeeper can limit what an agent can see, block certain actions, hide sensitive fields, log activity, apply rate limits, and require human approval before any changes are made.
Although the press releases will range from very select to rare, I said I'd pass...because sometimes the editors hide.
