Back
AnthropicClaude API
Claude Managed Agents permission policies now include auto: the server evaluates each agent or MCP tool call…
AI summary
Written by AI from the official notes. Check them for exact details.Claude Managed Agents now support automatic evaluation of tool calls, enhancing user control over permissions.
- New
autopermission policy evaluates agent tool calls automatically. agent.tool_useevents report evaluation results for transparency.- Version 1.32.0 adds
ant beta:sessions connectfor live session management.
Why it matters: Users managing Claude Managed Agents should care for improved control and monitoring of tool usage.
Full release notes2 changes
- Claude Managed Agents permission policies now include
auto: the server evaluates each agent or MCP tool call and runs it, denies it, or pauses for your approval.agent.tool_useandagent.mcp_tool_useevents report how each call was evaluated in anevaluationfield alongsideevaluated_permission. See Let the server evaluate each call withauto. - Version 1.32.0 of the
antCLI addsant beta:sessions connect, which attaches your terminal to a Claude Managed Agents session. You can follow the session live, send messages, and allow or deny tool calls that are waiting for approval. Pass--webto serve the Claude Console's session viewer locally and open the session there instead. See Connect to a Managed Agents session from your terminal.