Your developers are spinning up Claude Managed Agents right now. Autonomous agents that run code, browse the web, execute bash commands, and interact with external services — for hours, without human oversight. And they inherit the full permissions of whoever created them.
Security teams have no visibility into what those agents are doing.
Today, we're announcing Opsin support for Claude Managed Agents, extending our governance coverage to Anthropic's newest and most autonomous AI capability.
Claude Managed Agents is Anthropic's fully managed agent infrastructure, released last week. Instead of building a custom agent loop, developers spin up a session — an autonomous Claude instance running inside a managed cloud container — and point it at a task.
The agent can run shell commands, read and write files, search the web, fetch URLs, and connect to external services via MCP. Sessions run for minutes or hours. They execute multi-step workflows without waiting for human input.
This is not a chatbot. It is an autonomous actor that takes real actions inside real systems.
Claude Managed Agents run with the permissions of their owner. That means an agent created by a developer with access to production databases, internal APIs, and sensitive file systems can reach all of it — autonomously, over an extended session, without a human approving each step.
The risks are distinct from anything traditional AI security tools were built to handle:
Traditional DLP and CASB tools were built for data moving through predictable channels. An autonomous agent reasoning over your file system and executing multi-step workflows is not a predictable channel.
Opsin now supports Claude Managed Agents as part of our continuous AI governance coverage across Claude Enterprise, Microsoft Copilot, ChatGPT Enterprise, and Google Gemini.
With this support, security teams can:
Anthropic provides the infrastructure for running managed agents securely. What they don't provide is visibility into how those agents interact with your enterprise data — which documents they access, which APIs they call, what sensitive information moves through their sessions, and whether any of that aligns with your security policies.
That's the gap Opsin closes.
Claude Managed Agents is in beta and moving fast. Enterprises that wait to govern it will fall further behind every week new agents are deployed. The time to build visibility is before the blast radius grows — not after an incident makes it urgent.
Opsin's Claude coverage spans Chat monitoring, Projects, and now Managed Agents — the complete surface across every way Claude operates in your enterprise.
Your developers are spinning up Claude Managed Agents right now. Autonomous agents that run code, browse the web, execute bash commands, and interact with external services — for hours, without human oversight. And they inherit the full permissions of whoever created them.
Security teams have no visibility into what those agents are doing.
Today, we're announcing Opsin support for Claude Managed Agents, extending our governance coverage to Anthropic's newest and most autonomous AI capability.
Claude Managed Agents is Anthropic's fully managed agent infrastructure, released last week. Instead of building a custom agent loop, developers spin up a session — an autonomous Claude instance running inside a managed cloud container — and point it at a task.
The agent can run shell commands, read and write files, search the web, fetch URLs, and connect to external services via MCP. Sessions run for minutes or hours. They execute multi-step workflows without waiting for human input.
This is not a chatbot. It is an autonomous actor that takes real actions inside real systems.
Claude Managed Agents run with the permissions of their owner. That means an agent created by a developer with access to production databases, internal APIs, and sensitive file systems can reach all of it — autonomously, over an extended session, without a human approving each step.
The risks are distinct from anything traditional AI security tools were built to handle:
Traditional DLP and CASB tools were built for data moving through predictable channels. An autonomous agent reasoning over your file system and executing multi-step workflows is not a predictable channel.
Opsin now supports Claude Managed Agents as part of our continuous AI governance coverage across Claude Enterprise, Microsoft Copilot, ChatGPT Enterprise, and Google Gemini.
With this support, security teams can:
Anthropic provides the infrastructure for running managed agents securely. What they don't provide is visibility into how those agents interact with your enterprise data — which documents they access, which APIs they call, what sensitive information moves through their sessions, and whether any of that aligns with your security policies.
That's the gap Opsin closes.
Claude Managed Agents is in beta and moving fast. Enterprises that wait to govern it will fall further behind every week new agents are deployed. The time to build visibility is before the blast radius grows — not after an incident makes it urgent.
Opsin's Claude coverage spans Chat monitoring, Projects, and now Managed Agents — the complete surface across every way Claude operates in your enterprise.
