Claude Managed Agents get auto mode and a session viewer
Anthropic added two features to Claude Managed Agents on September 10, ClaudeDevs said on X. One is a session viewer in the ant CLI. The other is an auto mode in which Claude decides for itself whether each tool call should run.
The viewer comes in two forms. ant beta:sessions connect attaches your terminal to a session that is already running, and adding --web opens the same view as a web UI served from localhost. Managed Agents sessions can run on their own for hours, and their progress and outputs survive disconnections, so being able to attach in the middle of a run is the useful part. Before this, sessions could be inspected in the Console viewer, which was rebuilt for multi-agent sessions in August with one lane per agent and cost per thread. The latest ant CLI release on platform.claude.com is version 1.30.0, from September 3.
Auto mode changes who approves tool calls. With auto set, Claude checks each tool call against the intent you stated in user.message events, the messages you send the agent. Each check ends one of three ways: Claude runs the call, denies it, or stops and asks you for input.
Up next
Every operation in Anthropic’s threat report was disrupted
Anthropic disrupted every operation in its most detailed threat report yet, which covers attempts to push Claude into cyberattacks, influence operations, surveillance and biology. It calls them some of the most sophisticated misuse it’s seen.
Related stories
- Anthropic’s daily-brief agent treats a failed read as news
- ant apply syncs Managed Agents from files in a repo
- Memory reaches Self-Hosted Sandboxes in Managed Agents
- Claude Managed Agents now plans runs of up to 1,000 agents
- Anthropic hands Agent Skills to the foundation behind MCP
- Claude’s SDKs run the click loop, but the browser is yours
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.