Anthropic Unveils Claude Managed Agents to Streamline Enterprise AI Agent Development
Anthropic announced Wednesday the launch of a new product that aims to make it easier for businesses to build and deploy AI agents. The tool, Claude Managed Agents, offers developers out-of-the-box infrastructure to build autonomous AI systems, simplifying a complex process that was previously a barrier to automating work tasks.
This strategic launch puts Anthropic in a position to capitalize on its fast-expanding enterprise business. Just one day before the product announcement, on Tuesday, the company reported that its annualized recurring revenue has surpassed $30 billion, roughly three times the figure it posted in December 2025. Both Anthropic and OpenAI—its leading rival that already operates its own agent platform called Frontier—are in a race to build out robust enterprise-focused offerings as both companies prepare to go public as soon as this year.
Angela Jiang, head of product for Anthropic’s Claude Platform, shared that the vast majority of the company’s recent revenue growth comes from Claude Platform, the enterprise product that lets developers access Anthropic’s AI models via an API. Developers have already been leveraging Anthropic’s API to roll out AI agents such as Claude Code across their internal workspaces.
Jiang points out that there is a meaningful gap between what Anthropic’s AI models are technically capable of, and what businesses are actually able to build with them today. The new tool “enables any business to take the best-in-class infrastructure and deploy a fleet of Claude agents to do whatever work they need,” Jiang says.
Claude Managed Agents provides developers with a pre-built agent harness: the full stack of supporting software infrastructure that wraps around a core AI model to let it operate autonomously, or complete actions directly on a user’s behalf. In practice, this harness combines utility software tools, a dedicated memory system, and other core backend infrastructure. Agents built through the new service also come with a native sandboxed environment, where the agent can spin up new software projects in a secure, isolated setting. The product also lets developers build agents that can run independently for hours in the cloud, track activity from other deployed Claude agents, and adjust access permissions that control which internal or external tools an agent can use.
“When it comes to actually deploying and running agents at scale, that is a complex distributed-systems engineering problem,” says Katelyn Lesse, head of engineering for the Claude Platform. “A lot of customers we're talking about previously had a whole bunch of engineers whose job it would have been to build and run those systems at scale. Now that we are giving them that bit out of the box, they're able to have those same engineers be focused on the core competencies of their business and of their product.”
In a demo shared with WIRED, AI productivity startup Notion demonstrated how it is already using Claude Managed Agents to power a new automated client onboarding feature. Eric Liu, a Notion product manager, showed how he can offload an entire queue of routine client onboarding tasks within the Notion platform to a Claude Managed Agent, which then works through the task list one item at a time. While the agent-powered feature runs natively within Notion’s interface, Liu also pulled up a dedicated dashboard on the Claude Platform to show how teams can monitor agent performance and track which tools agents are accessing in real time.
Wall Street investors have grown increasingly wary of traditional software stocks in recent months, as Anthropic rolls out a steady stream of new enterprise AI offerings that many analysts argue could eventually make legacy software-as-a-service companies obsolete. Whether that threat ultimately comes to pass or not, the launch of Claude Managed Agents makes clear that Anthropic still has significant progress to make before most large enterprises fully run their operations on Claude.
