Anthropic brings Claude into Slack channels to join team conversations

Anthropic on June 23 put Claude Tag into Slack, letting employees summon the company’s AI agent into group threads by typing @Claude instead of opening a separate chat. The beta is available to Claude Enterprise and Claude Team customers and is designed to act like a visible participant in workplace conversations, not a private assistant hidden off to the side.

Admins can decide which channels Claude Tag can enter and which tools, data and codebases it can access. Once invited, the system can read the thread, break down tasks, remember relevant information over time and respond in-thread with what it creates. With ambient behavior enabled, Claude can also surface updates on its own and follow up on stalled discussions.

Instead of a single employee prompting a model in private, Claude Tag is built for shared channels where product managers, engineers and support staff can all see the same answer at once. Slack general manager Rob Seaman called the shift “making AI multiplayer,” and said Claude Tag appears in the open rather than in a private back-and-forth. Anthropic product leader Cat Wu said tagging Claude like a coworker is powerful. A use case connects it to Gmail so it can read messages, flag important senders and post alerts back into Slack.

Anthropic uses its own system as an internal benchmark. 65% of its product team’s code is created by its internal version of Claude Tag, and it already uses the tool for product metrics, support tickets and debugging. Claude Tag will expand to additional platforms in the coming weeks.