Claude Code artifacts can now call MCP connectors for live, viewer-specific data
Anthropic says Claude Code artifacts can now call MCP connectors, letting artifacts fetch information and take actions on demand using each viewer’s own connected accounts.
Anthropic says Claude Code artifacts can now call MCP connectors, enabling artifacts (dashboards, internal apps, reports) to fetch live data and take actions on demand — using each viewer’s own connected accounts (ClaudeDevs on X).
What’s new
According to Anthropic’s developer account:
- Artifacts can call MCP connectors, so you can build artifacts that fetch information and take actions for each viewer on demand (ClaudeDevs on X).
- This is available on Pro, Max, Team, and Enterprise plans (ClaudeDevs on X).
- It’s not available on publicly shared artifacts (ClaudeDevs on X).
Why it matters
Artifacts started as a way to publish a live, shareable page from a Claude Code session. Adding MCP connector calls pushes them closer to interactive internal tools:
- Build an artifact once, then let it pull live data when someone opens it.
- Because the artifact uses the viewer’s own connectors, creators don’t need to re-run sessions to refresh data — and viewers should only see the data they already have access to (ClaudeDevs on X).
What to watch
Anthropic’s post doesn’t list which connectors are supported or what permission model/approval steps apply. If you’re planning to ship an artifact to a broad team, confirm:
- Which MCP connectors are allowed in your org.
- Whether actions require additional confirmation.
- Whether connector calls are logged/audited.
Sources
- Announcement: https://x.com/ClaudeDevs/status/2077489907350856038