Anthropic launches ‘Claude Tag’ for team-based AI collaboration on Slack

Anthropic has introduced Claude Tag, a new feature designed to integrate its AI assistant directly into team workflows, starting with Slack. The tool allows organisations to treat Claude as a collaborative team member that can be tagged in conversations, assigned tasks and connected to internal tools and data.With Claude Tag, users can simply tag @Claude in Slack channels to delegate work. The AI then breaks down requests into steps, executes them using connected systems and shares results within threads. The feature is designed to work asynchronously, enabling teams to assign tasks and continue with other work while Claude processes requests.A key aspect of the rollout is its shared, multi-user functionality. Unlike traditional one-on-one AI chats, Claude operates as a single instance within a channel, allowing multiple team members to interact with it, track progress and build on previous tasks. This creates a more collaborative workflow, similar to working alongside a human teammate.Claude Tag is also built to retain contextual knowledge over time. By observing conversations in designated channels, it can recall relevant information and reduce the need for repeated instructions. With appropriate permissions, it can also draw insights from connected tools and data sources, though access is restricted and controlled to maintain privacy.Also read: WPP to pilot Meta’s AI creative tool, integrating it into WPP Open platformAnother feature is its ability to show proactive or “ambient” behaviour. When enabled, Claude can flag relevant updates, follow up on unresolved threads and surface useful insights without being explicitly prompted. This allows it to take a more active role in ongoing workstreams.The tool also supports long running and scheduled tasks, with Claude able to plan and execute work over extended periods, including hours or days. Teams can run multiple tasks in parallel, delegating different workflows to the AI simultaneously.Anthropic said Claude Tag builds on its earlier tools like Claude Code, making the system more proactive and better suited to team environments. Internally, the company noted that a significant share of its product team’s code is already generated using similar tools, with usage expanding beyond engineering into areas like customer support, data analysis and bug resolution.The initial launch is targeted at Claude Enterprise and Team customers in beta, with Slack as the first integration. Administrators can control access by defining which channels, tools and datasets Claude can interact with, effectively creating separate AI instances for different functions such as sales, engineering or support.To ensure governance, organisations can set usage limits, monitor activity logs and manage data permissions, with Claude’s memory and access restricted to approved environments.Anthropic plans to expand Claude Tag to additional platforms over time, positioning it as a broader system for team-based AI collaboration across workflows and business functions.