Anthropic launches managed agents for enterprise development

Anthropic has introduced Claude Managed Agents, a new tool aimed at helping enterprises build and deploy Artificial Intelligence agents more quickly by handling core infrastructure tasks. The release adds to Anthropic’s recent product push as it competes for a fast-growing enterprise market.

Anthropic has released Claude Managed Agents, a tool designed to help businesses build and deploy Artificial Intelligence agents with less engineering effort. The product provides out-of-the-box infrastructure for developers and, according to Anthropic, will enable enterprises to launch agents 10 times faster than before. The company says it takes over the complex pre-production work that normally delays production deployment, allowing customers to focus on defining an agent’s tasks, tools and guardrails while running the system on Anthropic’s infrastructure.

Anthropic said that pre-production work can include sandboxed code execution, checkpointing, credential management, scoped permissions, and end-to-end tracing. With Managed Agents, the platform automatically spins up a container for each agent, which can decide when to call tools, manage context and recover from errors. Anthropic also said the offering supports long-running sessions that can run for hours, while agent coordination for parallel work is available in research preview. Another research preview feature allows Claude to continually refine and self-evaluate response quality until it meets the user’s success criteria.

Managed Agents is currently available in public beta on the Claude platform. Anthropic identified users including Notion, Rakuten, Asana and others. Rakuten is using agents across its product, sales, marketing, finance and HR teams, with each deployed within a week. The company said removing the need to develop infrastructure has allowed it to focus more directly on democratizing innovation across the business.

Pricing for the new tool is based on consumption, with users paying for the models’ token use under Anthropic’s standard API costs, plus an extra $0.08 per session-hour for active runtime. The launch comes during a broader burst of releases from Anthropic as it competes closely with OpenAI in generative Artificial Intelligence. Earlier this week, Anthropic CFO Krishna Rao revealed that the company’s run-rate revenue had now surpassed $30 billion — more than tripling from $9 billion at the end of 2025 — and the debut of Managed Agents appears positioned to help Anthropic capture more demand from the evolving enterprise market.

52

Impact Score

Meta launches muse spark for its apps

Meta has introduced Muse Spark, an in-house large language model designed for its products and positioned as the first in a broader Muse family. The model brings multimodal reasoning, coding, shopping, and recommendation features to the Meta Artificial Intelligence app and website, with wider rollout planned.

Microsoft scales back Copilot in Windows 11 apps

Microsoft is pulling back some Copilot branding and interface elements from core Windows 11 apps after sustained user criticism. Notepad and Snipping Tool are among the latest apps to lose the prominent Copilot button as the company repositions some features.

Moderna rebrands cancer vaccine work as therapy amid federal skepticism

Moderna and Merck are increasingly describing an mRNA-based cancer vaccine as an individualized neoantigen therapy as vaccine skepticism reshapes the US policy environment. The shift reflects both scientific positioning and a broader effort to shield promising research from political hostility toward vaccines.

Uk business and trade committee scrutinizes Artificial Intelligence at work

The UK Business and Trade Committee has opened an inquiry into how Artificial Intelligence is reshaping the workforce and whether existing workplace protections remain adequate. Employers face rising pressure to improve transparency, fairness, oversight and data governance as regulators intensify scrutiny.

Anthropic launches Project Glasswing for cyber defense

Anthropic has introduced Project Glasswing to address mounting cybersecurity risks tied to increasingly capable Artificial Intelligence models. The initiative brings major technology and finance companies together to use Claude Mythos Preview as a defensive tool for critical software.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.