Anthropic's Claude Code Artifacts update brings live, shared dashboards and interactive workspaces to enterprises

Anthropic announced a potentially game-changing new feature for users of Cloud Code on Cloud Team and Enterprise subscription plans: Artifacts.

This update turns the Cloud Code session work into a live, interactive, and shareable, custom HTML webpage, allowing a Cloud Code user to plug in live code, multiple data sources, and have it displayed on an interactive URL that they can send to other teammates – whether it’s a dashboard, an app design, or any other product for internal use.

These teammates and end users can see the webpage updating in real time as the cloud code does its work autonomously or under the guidance of the user, and as the connected data sources and codebase change.

While Anthropic first introduced Artifacts to its consumer web chatbot in the summer of 2024 – where it evolved from a manual toggle feature into a generally available tool for publishing code snippets and games on the web – integrating this capability directly into cloud code command-line interfaces (CLI) and desktop apps bridges the gap between deep, back-end engineering and the non-technical stakeholders who need to understand it.

Products and Technology: End of Status Update

At its core, cloud code artifacts act as a dynamic translation layer. Built directly from the unbroken context of the user’s session, the agent uses a local repository codebase, connected monitoring tools, and conversational logic to spin up specialized web pages.

Engineers no longer need to wire up external data sources or erect temporary infrastructure; AI builds on pre-existing UI.

The important thing is that these web pages are not static exports. As the AI ​​works through a terminal session, the open webpage is refreshed in place, updating charts and text immediately to the same URL. Each update publishes a new version history, allowing teammates to securely revert or track the agent’s progress on desktop or mobile.

The battle of live, interactive, shared AI work surfaces: Anthropic’s Cloud Code Artifact vs. OpenAI’s Codex sites

Anthropic’s update comes more than two weeks after OpenAI released a massive update to its own Codex platform, which introduced a substantially similar enterprise hosting feature called "sites".

This tit-for-tat product cadence highlights the increasingly escalating battle over enterprise scope between functions and beyond the developers themselves, though some important technical and philosophical distinctions should be made for enterprises to consider.

As stated in their respective developer documentation webpages, OpenAI is building a platform as a service; Anthropic is creating a stateless canvas.

OpenAI’s sites are designed to generate sustainable, full-stack web applications. According to the platform’s documentation, Codex sites host projects that output as Cloudflare Worker-compliant ES modules.

Importantly, sites support persistent backend infrastructure: agents can automatically wire up "D1" Relational databases for structured data (such as user progress or saved records) and "R2" Object storage for file uploads. An OpenAI site can support public sign-in, integrate with external identity providers, and allow highly specialized access controls tailored to specific workspace groups.

It uses a two-step publishing process – saving a reviewable candidate associated with a Git commit before officially deploying it to production. In short, it is a production environment designed to replace functional internal SaaS tools.

In contrast, Anthropic’s cloud code artifacts deliberately avoid the backend. The newly released document is clear about its limitations: "An artifact is a capture of work, not an application".

Each artwork is a single, self-contained HTML page 16 MiB in size. To guarantee organizational security, the cloud wraps the published file in a strict Content Security Policy (CSP) that blocks all external network requests. Tea

This means that the page cannot load external scripts, fonts, or stylesheets, and fetchXHR, and WebSocket calls are completely blocked. All CSS and JavaScript should be inline, and images should be embedded as data URIs. Artifacts cannot store form inputs, call APIs at view time, or serve multiple routes.

This technical limitation is actually Anthropic’s deliberate philosophical position: while OpenAI wants to continuously spin up software portals for the entire company, Anthropic is keeping cloud code firmly rooted in short-term, highly secure technical workflows. are cloud artifacts No Meaning of having software; They’re meant to replace whiteboard diagrams, manual bug walkthroughs, and status reports with secure, self-updating visual tools that never leak live data outside the corporate boundary.

Licensing and enterprise security: keeping the codebase private

Because these agents sit at the nexus of proprietary company data and live codebases, licensing and access control is a primary concern.

Both Anthropic and OpenAI have opted for closed, proprietary licensing models for these new visual workspaces. For end users and developers, the difference is significant. Unlike permissive open-source software (such as MIT or Apache 2.0) or strict copyleft licenses (such as the GPL)—which provide developers with legal freedom to inspect, modify, and self-host the underlying code—neither cloud code artifacts nor codex sites can be freely forked or hosted on the cloud.

Enterprise clients do not maintain code-level ownership over Anthropic’s rendering engine or Codex’s integration nodes; Both work strictly within their limits Infrastructure managed by the respective creators.

To make this vendor-managed approach convenient for enterprise compliance teams, both companies have placed a heavy priority on organizational security. Anthropic ensures that each artwork is private to its author by default and cannot be made public on the wider Internet. When an engineer chooses to share a link, it is viewable exclusively by authenticated members of their specific organization. System administrators maintain ultimate authority, managing access through org-level toggles, role-based scoping, and clear retention policies, while maintaining oversight through a centralized compliance API.

OpenAI takes a similar gated approach with Codex Sites, rolling out the feature primarily to ChatGPAT Business and Enterprise Workspaces. Like Anthropic, OpenAI relies on system administrators to manage deployments through centralized workspace settings, which requires the administrator to explicitly enable sites through role-based access control (RBAC) for enterprise tiers.

However, because Codex sites function more like hosted web applications, its access controls are a bit more detailed. When an engineer is ready to share a deployed URL, they can apply specific access modes: restricting the site to only themselves and workspace administrators, opening it to all active users in the workspace, or limiting access to custom user groups.

Additionally, to prevent sensitive data leaks, OpenAI provides a dedicated site panel to securely manage runtime environment variables and secrets, ensuring that those keys do not have to be committed to local source files.

Reactions and Thoughts

The introduction of visual, self-updating UI layers for command-line agents is fundamentally changing how developers view their own workflows. As AI handles raw syntax and automates reporting, the hassle of communicating technical work to stakeholders is disappearing.

Cloud Code head and creator Boris Cherny highlighted the broader utility of the update in a post on X earlier today:

"I’m using Artifacts in Cloud Code for everything: visual explanations of tricky code, system diagrams, quick previews of certain animation options, data analysis, and dashboards that I share with the team," Cherney wrote. "They are a game changer for how I work with the cloud. Can’t wait to hear what you think!"

This sentiment is practically demonstrated in Anthropic’s launch materials. In one scenario, an engineer prompts the cloud code to investigate user drop-off since the previous software release.

In a matter of seconds, the agent performs a SQL read, creates an interactive drop-off funnel dashboard, and diagnoses "Pro accounts stop on export sheet". The AI ​​then proposes UI improvements, updates the live chart as soon as the code is reactivated, and generates a secure link that the manager can instantly open via mobile.

By turning the terminal into a living, collaborative canvas, Anthropic is proving that the most valuable output of an AI coding assistant isn’t just the code itself – it’s the context, logic, and ability to instantly share that work.

<a href