Top news
Investing in the public benefit OfficialAnthropic NewsOct 18Popular Musk Pledges SpaceX Shares to ‘Every Kid in America’ WebBloomberg TechnologyOct 19Popular Websites are closing their doors to AI agents WebThe Verge AIOct 17 This is the new ChatGPT Voice, powered by GPT Live OfficialOpenAI on YouTubeOct 18 How AI giants will reshape global philanthropy WebAxios AIOct 19 Gemini 4 Argon Really Surprised Me WebTwo Minute PapersOct 19 Google Gemini 4 agents just beat Minecraft with no human intervention WebEngadget AIOct 19 Who could it be? 🧐 OfficialGoogle on YouTubeOct 18 Anthropic gives outside AI watchdog “employee-like” access WebArs Technica AIOct 18 What are self-replicating ‘AI worms’, and why is Anthropic worried about them? WebThe Guardian AIOct 18
GlossaryAgentsIntermediate

Managed_agents

Also written: AI coworkers, Internal agents, Managed Execution Layer, Agent-supported processes, Team agents, Claude Managed Agents, Copilot agents, AgentCore companion

Managed agents are AI systems controlled and operated within a specific platform, simplifying deployment and task automation.

In plain terms

Managed agents are AI systems designed to perform tasks automatically. Think of them like specialized workers within a larger AI system. A platform manages these agents, making it easier to deploy them and control what they do.

For example, Anthropic has added cookbooks to help developers build "managed agents" that can handle tasks like reviewing budgets or acting as advisors. These agents are built and run within Anthropic's framework, simplifying the process for users.

The expert version

Managed agents refer to AI agents that are controlled, orchestrated, and operated within a specific framework or platform. This management layer typically handles aspects like deployment, execution, monitoring, and potentially resource allocation or access control. These agents can be designed to perform multi-step workflows, interact with external tools or APIs, and operate with predefined capabilities.

Platforms offering managed agents aim to abstract away the complexities of agent development and deployment, allowing users to focus on defining agent behavior and objectives. This often involves providing SDKs, APIs, or user interfaces for agent creation and management, along with features like environment hooks, budget controls, and integration with other services.

Why it matters now

Companies like Anthropic and Google are actively developing and documenting systems for managed AI agents. These systems aim to simplify the creation, deployment, and security of AI agents that can perform complex, multi-step tasks, often within controlled environments.

Where it shows up 24 items, newest first
How each company uses it
Anthropic Claude
Anthropic's documentation and cookbook examples show a focus on building secure AI agents, simplifying workflows, and organizing agent applications for developers.
Google Gemini
Google's Antigravity Agent automates complex multi-step tasks securely within a sandbox environment.
OpenAI ChatGPT
OpenAI is pushing advancements in agents alongside models and hardware, indicating a broader focus on agent capabilities.
Related terms

AI agentsAutonomous agentsAgent frameworksOrchestrationTool useLLM applications

Often confused with
Autonomous agents
While managed agents can be autonomous, the term 'managed' em
AI models
Managed agents are typically built *using* AI models, but they repres
APIs
APIs are interfaces for interaction, whereas managed agents are the operat
Origin

The concept of managed agents emerges from the broader field of AI agent development and the need for more robust and user-friendly ways to deploy and control them. Specific implementations and terminology are being defined by leading AI labs as they build out their platforms.

Try this about 5 minutes

Explore the Anthropic Cookbook or Google Gemini API documentation for examples of managed agents. Look for tutorials on creating an agent that can perform a specific task, like summarizing a document or answering questions based on provided context, within their respective platforms.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

What is the main benefit of managed agents?
Managed agents simplify the process of building, deploying, and controlling AI agents, allowing users to focus on their tasks rather than the underlying infrastructure.
Can managed agents operate independently?
Yes, managed agents can be designed to operate autonomously to complete tasks, but they function within the operational and control framework provided by the managing platform.

First seen on Subvolts Nov 13, 2025. Updated Oct 17, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.