Agentic AI Platform Change Log

Browser/computer use

Claude in Chrome is now generally available for all paid Claude plans. The extension can now take actions autonomously in the browser without requiring user approval for every step, relying on a safety classifier to validate each action before execution.

General availability signals enterprise readiness and stability for organizations using paid tiers. The addition of autonomous execution removes a major friction point in multi-step web workflows, enabling true delegation of browser-based tasks.

Memory / state

Anthropic introduced cross-surface memory for Claude Cowork, which powers the Claude in Chrome side panel. Users can now manage, edit, or delete remembered topics, with sensitive topics excluded by default.

Persistent memory allows the browser agent to retain context across different sessions and devices. This improves efficiency for users conducting ongoing research or repetitive web tasks, as they no longer need to re-prompt the agent with their preferences.

Browser/computer use

Anthropic updated the Claude in Chrome extension to operate as a full Claude Cowork session, enabling session synchronization across desktop, web, and mobile platforms. The update automatically integrates a user's skills and connectors into the browser environment and adds a mandatory human-approval step before the agent executes high-risk actions like submitting forms.

Users can now start complex workflows in a browser tab and continue them on other devices without losing context or history. The introduction of mandatory user confirmation for sensitive actions provides critical oversight, significantly reducing the risk of prompt injection attacks from untrusted websites.