Internet & Software Tips

Code Claude adds support for artifacts – SD Times

A Claude Code session can range from investigating an incident to re-engineering a service to analyzing months of data. Artifacts translate work into a web page that anyone can open and inspect, such as a pull request flow, a dashboard that you can filter and organize, or even a release checklist that populates as work is done. Artifacts make it easy to collaborate on shared work, so teams can spend more time building and less time talking about status updates.

It is built on top of the context from your session

Code Claude creates an artifact using the full context of your session, including your codebase, your connectors, and the conversation itself. A single incident page can include failed tests and work-behinds from your code, error escalations from a connected monitoring tool, and causal reasoning from the session you just ran. With artifacts, you don’t need to connect data sources or set up infrastructure. You request a page, and Claude’s Code builds on what’s already there.

Live pages update locally

When Claude Code updates an artifact, the open page restarts locally and colleagues see the updates as they are published. Every publication is a new version in the same link, with a version history so you can restore at any time, and the gallery allows you to browse and manage all the artifacts you have made.

From our internal testing, one of our most common use cases has been debugging. This usually looks like: The engineer starts investigating the incident before standing up. Claude’s code runs through the logs and publishes the artifact: timeline, suspect performance, and error rate chart. He shared a link with his team from the header of the page. By the time the stand-up started, Claude had republished it twice as the investigation progressed, incorporating more recent information. With artifacts, team members and stakeholders don’t have to “redirect us to the agent’s findings” because they are all looking at the same point of view, in the same context.

It is confidential to your organization

Every artifact is private to its author by default. When you’re ready, share with your colleagues and your organization directly from the page. Artifacts are only visible to verified members of your organization and cannot be made public. Administrators manage access with org-level toggle and role-based scope, set retention policies, and gain global visibility through the compliance API.

Getting started

Ask your session for an artifact – or just ask for something visual, hHere are some thoughts on the role:

  • Official / open source: License check for all dependencies, marking the remaining copy, directly from the repo. “Create an artifact that lists all third-party dependencies and their license, flagging anything that’s copied.”
  • Privacy: A data flow map where personal data is collected, stored, and coded. “Track where we touch personal data across the codebase into a privacy review artifact.”
  • Security: Found connected to a straight line, so the correction is not clear. “Create an auth discovery artifact from this review, each associated with the code.”
  • FinOps / financial platform: Cloud resources and cost drivers mapped from your infrastructure as code. “Map our cloud resources from Terraform to artifact, service-aggregated, with significant cost drivers.”
  • Software engineers: PR or bug walkthrough reviewers can actually follow, removed from diff and surrounding code. “Make an artifact to walk in this PR – the difference, the thinking, and what I have tested.”
  • Leading designers and engineers: Several UX directions for the screen, each of which is built with your own elements so that the one you choose can be deployed. “Give me an artifact with 5 UX types for this registration form, built with our component library.”
  • Civil engineers and architects: A map of how the service actually fits together, drawn from a real import graph instead of a whiteboard. “Map how a payment service integrates into an artifact, from code.”
  • SRE & on-call: An incident page that expands as you scroll and becomes a postmortem. “Make this incident an artifact – the timeline, the suspect’s actions, the error from our surveillance – and republish it as I work.”
  • Engineering managers: A page of what is actually posted, built from aggregated PRs. “Create an artifact of what was assembled in my team this week from the PRs, collected by the project.”

Claude’s code builds the page and gives you the link. Open it in your browser or desktop app, share it from the header—updates are published to the same URL automatically.

Availability

Artifacts is available in beta to Claude Team and Enterprise orgs, from the Claude Code CLI and desktop application, with pages viewable in any browser.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button