Claude Code Now Supports Artifacts: Turn Coding Sessions Into Live, Shareable Web Pages
Anthropic introduced Artifacts in Claude Code, a new beta capability that converts an active coding session into a live, shareable browser page. Available for Team and Enterprise organizations, Artifacts auto-update as Claude works: letting engineers share incident pages, PR walkthroughs, dashboards, and release checklists with teammates in real time. Access is restricted to authenticated organization members by default, with admin-level controls for retention and sharing policies.
Sources & Mentions
5 external resources covering this update
Anthropic's Claude Code Artifacts update brings live shared dashboards and interactive workspaces to enterprises
VentureBeat
Anthropic brings Artifacts to Claude Code, letting teams share live pages from coding sessions
The Decoder
Anthropic brings live, shareable Artifacts to Claude Code
DevOps.com
Claude Code Artifacts turns AI coding sessions into live, shareable browser pages
TechTimes
Claude Code: Hooks, Subagents, and Skills Complete Guide
Dev.to
Claude Code Gains Artifacts: Live, Shareable Session Pages for Teams
Anthropic launched Artifacts in Claude Code on June 18, 2026, giving engineering teams a way to transform an AI coding session into a self-contained, browser-viewable page that updates automatically as work progresses. The feature is available in beta for Team and Enterprise organizations through the CLI and desktop app.
What Artifacts Are
An artifact is a live HTML page generated directly from a Claude Code session's full context: the local codebase, connected monitoring tools, conversation history, and any active data sources. As Claude Code modifies the artifact, the page refreshes automatically for all viewers. Previous versions are preserved in a gallery with full restore capability.
Use cases include:
- Incident pages: Security engineers can create findings linked to specific lines of code; SREs can generate incident pages that automatically evolve into postmortems as the session progresses.
- PR walkthroughs: Engineers can produce a navigable summary of a pull request, with the AI's analysis embedded alongside the relevant code.
- Dashboards: An engineer can prompt Claude Code to investigate a user drop-off issue, generate an interactive funnel chart, propose UI fixes, and update the live artifact in real time as refactoring proceeds.
- Release checklists: Items fill themselves out as the session's work gets done.
How to Use It
To create an artifact, users simply ask for one, or for anything visual, during a session. Claude Code returns a shareable link that can be opened in a browser or the desktop app. The link is surfaced directly in the session header for quick distribution.
Privacy and Access Controls
Artifacts are private to their author by default. Sharing options are scoped to team members and organizational colleagues; only authenticated organization members can view shared artifacts. Enterprise administrators have access to retention policies and access management controls.
Each artifact runs under a strict Content Security Policy that blocks all external network calls, defining a deliberate security boundary.
What's Coming
Anthropic has committed to keeping Artifacts in beta through Q3 2026. The published roadmap includes link-level permissions (restricting viewing to specified email domains), permanent artifact pins, and an embeddable iframe option for integration with tools such as Confluence or SharePoint.