Tony Kim
Jun 09, 2026 21:28
Claude Managed Agents now support scheduled tasks and secure CLI tool integration, streamlining enterprise AI automation.
Anthropic has expanded the capabilities of its Claude Managed Agents, announcing two major updates: scheduled task automation and secure integration with command-line tools. Both features are now available in public beta as of June 9, 2026, marking another step forward in simplifying enterprise AI workflows.
Scheduled Task Automation
The new scheduling feature allows Managed Agents to execute recurring tasks without requiring users to develop or host custom schedulers. Teams can assign agents a cron-style schedule to automate repetitive processes like nightly data syncs, weekly compliance scans, or daily activity summaries. Users retain full control, with options to pause, resume, or trigger tasks on demand.
Notable early adopters include:
Rakuten, which automates data analysis and reporting while monitoring application performance without additional dashboards.
Actively AI, which uses scheduled deployments for automated cross-account searches, replacing custom scheduling infrastructure.
Ando, where agents autonomously track tasks, send reminders, and follow up, supporting hiring and sales teams.
Secure CLI Tool Access via Vaults
The second update extends Claude’s vault functionality to securely store and manage environment variables, enabling authenticated API and CLI tool access within an agent’s sandbox. Developers can register API keys tied to specific domains, ensuring secure, controlled integration with external systems. Importantly, agents never directly access these credentials; keys are applied only at the network boundary for approved requests.
This feature is designed to streamline integration with tools like Notion, Browserbase, and KERNEL. For example:
Notion uses vault-stored environment variables to securely roll out its CLI capabilities, adding file-upload functionality without exposing API tokens.
Browserbase leverages the feature to automate validation of its browser skills catalog.
Milana connects AI agents to customer codebases, enabling autonomous bug fixes and large-scale data analysis.
Strategic Implications for Anthropic
These updates align with Anthropic’s broader strategy of positioning Claude Managed Agents as a turnkey solution for deploying long-running, stateful AI workflows. Launched in April 2026, the service has gained traction among enterprises by offloading the complexities of orchestration, sandboxing, and execution to Anthropic’s infrastructure. Other recent enhancements, such as persistent memory and event streaming, further underscore the platform’s focus on scaling AI automation for enterprise use cases.
However, it’s worth noting that while these features enhance functionality, Claude Managed Agents remain ineligible for Zero Data Retention or HIPAA compliance due to server-side state storage. This could limit adoption in privacy-sensitive industries like healthcare.
Looking Ahead
As enterprises increasingly seek to integrate AI into production environments, features like task scheduling and secure CLI integration could drive broader adoption of Claude Managed Agents. With Anthropic’s continued focus on enterprise partnerships and new tools, the platform appears well-positioned to capture demand in the growing autonomous AI market.
Image source: Shutterstock
