Claude Tag Launches in Slack: @Claude Becomes Your Team's Shared AI Teammate

TL;DR: Anthropic launched Claude Tag today — @Claude in any Slack thread summons a shared agentic teammate that plans tasks, reads channel context, and proactively flags updates across your whole team.

Claude Tag, now in research preview for Enterprise and Team customers, is Anthropic's first persistent Slack agent. Type @Claude in any thread:

• Execute multi-step tasks independently. Claude reads the conversation, breaks the request into stages, completes them, and delivers the result to the team — not just to the person who asked.
• Proactive monitoring. Tell Claude to watch a channel and surface posts that match a condition ("flag anything that touches the payment API") — it does so without being re-prompted.
• Shared identity. Every member interacts with the same Claude "identity." Half-finished tasks can be handed off; context is shared across the team, not siloed to individuals.

Anthropic says the Claude Code team has used Claude Tag internally all year. It now writes 65% of the product team's code, including most of what built Claude Tag itself.

Note: June 23 outage. Earlier today the API, Claude.ai, Claude Code, and Claude Cowork experienced two 529-error (overloaded) incidents, peaking at 8,000+ Downdetector reports. Both were resolved within an hour; Anthropic has not published a root-cause analysis.

Why it matters: Claude Tag is Anthropic's push to make Claude a visible team participant in the tool every engineering team already lives in — not just a terminal tool for individuals. If your team uses Claude Code, this makes Claude's work visible and handoff-able in Slack without a context switch.

Sources: TechCrunch: Claude Tag, Fortune: Claude Tag launch, Engadget: Claude in Slack, Claude status history