Anthropic’s Claude Tag learns your company through Slack messages
▼ Summary
– Anthropic is introducing Claude Tag in research preview, an “always-on Claude” that lives in Slack and acts as an AI teammate, available for Claude Enterprise and Claude Team customers.
– Claude Tag adds persistent context and memory, learning about the work as it follows channels and can automatically gather facts from other channels if permitted.
– Everyone in a Slack channel can access a single Claude identity, allowing them to see Claude’s work and continue conversations, with system administrators controlling access to tools, information, and channels.
– When assigned a task, Claude Tag breaks it into stages and works through them with available tools, and it also has an ambient mode that proactively updates the team and follows up on forgotten tasks.
– Anthropic’s focus on organizational context is shared by competitors like Microsoft (Copilot/Work IQ), Snowflake, Databricks, and Glean, which are building systems to tap into tacit company knowledge.
Anthropic has unveiled Claude Tag in research preview, positioning it as an “always-on Claude” that functions as an AI teammate directly within Slack. This new feature allows users to tag @Claude for real-time insights in chats and to delegate tasks, launching first for Claude Enterprise and Claude Team customers through their existing Slack integrations.
This release builds on earlier capabilities where users could already DM @Claude or mention it in channels for on-demand assistance. Similarly, Claude Code in Slack has been handling coding tasks by routing channel mentions to full web-based coding sessions, then posting results back into the thread. However, Claude Tag introduces a more persistent layer of context and memory that previous tools lacked.
“Claude follows along with its channel, learning ever more about the work,” Anthropic explains. “It can also automatically gather facts from elsewhere in the organization, if granted permission to read other channels.” This means every member of a Slack channel accesses a single Claude identity, allowing anyone to see what the AI has been working on and continue conversations where others left off.
System administrators control which tools, information, and channels Claude can access. Each Claude identity remains strictly scoped to the channels admins define, preventing a legal-focused Claude from seeding memories into an engineering channel, for instance. When assigned a specific task, Claude Tag breaks it into stages and works through them using available tools, responding in a Slack thread with its output.
Claude Tag also includes an ambient mode that proactively joins conversations to keep teams updated, flag relevant information from across the organization, and follow up on forgotten threads or tasks. Anthropic says this creates the feeling of “working with a real colleague , one that can produce work in public view, with far greater context and understanding than before.”
That contextual awareness is becoming crucial for enterprise deployments, and Anthropic is not alone in targeting it. Microsoft leverages its Graph through Copilot and Work IQ. Snowflake and Databricks are positioning their platforms as back-end repositories of tacit organizational knowledge that agents can draw upon. Glean is also building an intelligence layer that sits between the model and enterprise data, understanding company context to deliver more relevant responses.
(Source: TechCrunch)
