OpenAIAnthropicGooglexAIPerplexity
← Back to all updatesPublished ·Anthropic• Minor
Claude Managed Agents now supports setting effort in an agent's model configuration.
The read
Effort is now a first-class setup choice for Claude Managed Agents.
The frame
It sits inside the model object, keeping effort tied to model configuration when creating managed agents.
Primary source
Anthropic →Keep reading
See what Nova3 builds →Related updates
- Aug 12, 2026Claude Managed Agents webhooks now cover environment and memory store lifecycle events without polling.
- Aug 12, 2026`version` is now optional when updating a Claude Managed Agents agent.
- Aug 12, 2026Anthropic is sunsetting the legacy Claude Console Workbench, with access ending on August 17, 2026.
This is the world we build in. It moves this fast. We move with it.
Bring us your project →