Anthropic launches Claude Tag, a persistent Slack team member powered by Claude Opus 4.8

Anthropic today launched Claude Tag in beta, a new product that adds Claude as a persistent member of Slack workspaces. Unlike existing point-in-time AI integrations, Claude Tag gives every member of a Slack channel a shared AI presence they can address by typing @Claude — one that builds context over time as it observes the channel's ongoing work.

What's new

Claude Tag is available starting June 23, 2026 to Claude Enterprise and Team customers. The integration runs on Claude Opus 4.8 and works directly inside Slack, where a single Claude instance serves everyone in a given channel.

Key capabilities:

• Multiplayer by design. "Within a given Slack channel, there's one Claude that interacts with everyone." Any team member can tag Claude, delegate a task, or ask a question without switching tools.
• Contextual memory. "As Claude follows along with its channel, it builds more context about the work," reducing the need to re-explain background on every request.
• Ambient mode. When enabled, Claude proactively surfaces information it thinks the team might need — without being explicitly asked.
• Asynchronous task execution. "Set Claude a task, and you can focus on your other priorities while it works. It can also schedule tasks for itself," enabling multi-hour workflows that run in the background.

Anthropic reports that their internal deployment has had a measurable impact on engineering output: "65% of our product team's code is created by our internal version of Claude Tag."

Context

Claude Tag follows a wave of AI assistant integrations that have attempted to embed language models into team collaboration tools. Earlier efforts — bots that respond to slash commands or generate single-turn answers — have generally required users to context-switch or re-explain their situation each time.

The key architectural difference with Claude Tag is persistence. Rather than treating each @mention as a fresh conversation, Claude Tag maintains state across a channel's history, allowing it to understand evolving project context, track prior decisions, and adjust its assistance accordingly. The ambient mode takes this a step further, enabling proactive communication without an explicit prompt.

Anthropic positions this as part of a broader push toward agentic AI that works alongside human teams rather than responding on demand. Claude Opus 4.8, which powers Claude Tag, was released in early 2026 and is Anthropic's current flagship model for complex, long-horizon reasoning tasks.

Why it matters

Claude Tag shifts AI from a lookup tool to a team member. For engineering and product teams in particular, having an AI that already understands a channel's decisions, codebase discussions, and ongoing priorities removes a significant amount of overhead from every individual interaction.

The 65% code-creation figure from Anthropic's own product team is notable, though it applies to an internal version built by the team that also develops the underlying model — a best-case deployment. Still, it signals Anthropic's confidence in using Claude Tag as a genuine operating layer, not just a demo.

For enterprise customers on Team and Enterprise plans, Claude Tag represents a meaningful expansion of what they can do with Claude subscriptions they already pay for. Whether organizations adopt ambient mode broadly will likely depend on how well it learns to distinguish signal from noise in active Slack environments — a problem that scales with channel volume and topic breadth.