Anthropic announces Claude Managed Agents, offering developers an agent harness and other tools to build and deploy AI agents at scale, available in public beta
Amid rapid enterprise growth, Anthropic is trying to lower the barrier to entry for businesses to build AI agents with Claude.
WiredMaxwell Zeff
Context & Ripple Effects
Anthropic has been assembling the layers for more autonomous software work: API primitives such as code execution and MCP connectivity, then parallel-coordinating agent teams in Claude Code. Managed Agents moves that stack from individual features toward a deployable developer harness.
The later addition of scheduled memory review for Managed Agents suggests the product is being extended beyond one-off task execution toward agents that retain and improve operational context over time. That makes the launch consequential for enterprise teams that need repeatable deployment patterns rather than bespoke agent prototypes.
First-order effects
Developers can use Anthropic’s public-beta harness and tools to build and deploy Claude-based agents with less infrastructure work, concentrating more of the implementation workflow inside Anthropic’s platform.
Anthropic expands its Claude offering from model access and point features into managed agent operations, creating a clearer path from experimentation to scaled deployment.
Second-order effects
Companies building agent applications may reassess whether to maintain their own orchestration and deployment layers or adopt Anthropic’s managed stack; competing model platforms face pressure to offer similarly integrated agent tooling.
The value of Anthropic’s earlier agent-building API capabilities rises when those components can be used within a managed deployment workflow, increasing the importance of ecosystem compatibility and developer workflow integration.
Third-order effects
If managed harnesses become the default enterprise route to agent deployment, competition will shift from model quality alone toward reliability, memory, orchestration, and controls across the full agent lifecycle.
This could favor vendors that can pair agent platforms with implementation capacity, while leaving enterprises to decide how much operational control to trade for faster deployment; the outcome depends on whether managed systems meet production requirements.
The trend: AI vendors are productizing the operational layer around foundation models, turning agent creation from a collection of developer primitives into managed enterprise platforms.
Shipping a production agent meant months of infrastructure work first. Managed Agents handles that for you. Define your agent's tasks, tools, and guardrails, and we run it on our infrastructure. Here's what early customers have built: [image]
Before, you had two ways to use @claudeai with Notion: 1. Custom Agents (Notion runs the harness) 2. Claude + MCP (Claude runs the harness) Now there's a third with Claude Managed Agents. Notion is the collaboration layer. You can keep working in Notion while Anthropic runs the
Managed Agents is the first ‘agent in the cloud’ API that has the right mix of simplicity and complexity. Implementation details like how you manage a sandbox are abstracted, but you have a lot of control over the actual execution of the model.
Introducing Claude Managed Agents: everything you need to build and deploy agents at scale. It pairs an agent harness tuned for performance with production infrastructure, so you can go from prototype to launch in days. Now in public beta on the Claude Platform. [video]
@Rakuten deploys specialist agents, powered by Managed Agents, for product, sales, marketing, and finance, each one in under a week: https://claude.com/...
Anthropic's new Claude Managed Agents is the next shit from AI that simply responds to AI that can actually do the work. They are on the right path, updates being shipped daily, no end in sight. So far it feels like 2026 is Anthropics year.
a new addition to the Claude API: Claude Managed Agents. you supply the agent configuration and task, we handle to agent harness + managed infrastructure to ensure it works reliably over long task horizons. https://x.com/... [image]
Anthropic just mass-obsoleted every agent orchestration startup in a single launch. The screenshot tells the full story. That's a production fleet dashboard. 8 agents running. 247 completed tasks. Active status. MCP-connected to HubSpot, pulling deals, generating proposals,
today we're releasing Claude Managed Agents in public beta. long-running, autonomous agentic systems are the future. and our new suite of apis means engineering teams can harness Claude's exponential power with scalable infra out of the box. we can't wait to see what you build
@asana built AI Teammates on Managed Agents: agents that work inside Asana as part of your team, picking up assigned tasks. Offloading the infrastructure let them put engineering time into the multiplayer experience.
Anthropic banned OpenClaw from using Claude subscriptions 4 days ago Today they just launched their own managed agents platform So basically OpenClaw without all the headache and cheaper usage
I just pulled an all-nighter building exactly this. Have never been more excited in my life. Didn't eat. Didn't sleep. Cancelled meetings. Thought I was a genius. Wake up and this is the first tweet I see. Not gonna lie, feeling a bit of despair. Of course I have no reason to
This is a business in a box. Claude just made it possible to build and deploy AI agents at scale with managed infrastructure. Here's how to sell it: 1. Pick a niche (real estate, e-com, coaching, agencies) 2. Build one agent that solves their #1 repetitive problem 3. Deploy it
Agent = model + harness Managed Agents = agent + runtime + infra (fully hosted) Anthropic wants to sell agents, not only the models. It's a huge market, and it will change the pricing structure away from tokens. (They ship so fast because they have Mythos. I want it so much.)
I've found Managed Agents to somehow be both the fastest way to hack together a weekend agent project and the most robust way to ship one to millions of users. It eliminates all the complexity of self-hosting an agent but still allows a great degree of flexibility with setting