AI Models

Claude Cowork Now Runs Inside Chrome Extension Side Panel

Anthropic has integrated Claude Cowork into the Chrome extension's side panel, enabling users to create Excel files, PowerPoint slides, and reports directly in the browser. The update, available for paid plans, requires no extra setup and supports cross-device sessions. Security warnings remain about prompt injection risks and sensitive data handling.

Neura News

Neura News

Neura Market Editorial

August 13, 20263 min read
Claude Cowork Now Runs Inside Chrome Extension Side Panel

Anthropic has integrated Claude Cowork directly into the side panel of its Chrome extension, letting the AI agent create Excel files, PowerPoint slides, and reports without leaving the browser. The update, available now for all paid plans through the Chrome Web Store, marks the first time skills, plugins, and connectors work in the browser with no extra setup.

What Changed in the Extension

The Chrome extension previously existed for browsing and gathering information. It could read the page you're logged into, click, type, and fill out forms. Now, a full Claude Cowork session runs inside the panel, turning the browser into a workspace rather than just a research tool.

Claude in Chrome handles the research side of a task. Cowork then takes that material and transforms it into finished deliverables. Anthropic lists several use cases, including pulling metrics from analytics dashboards, organizing Google Drive, and logging sales calls in Salesforce.

How the Workflow Fits Together

The division of labor is straightforward. Claude reads pages and collects data while you browse. Cowork then builds the output, whether that means generating a spreadsheet, assembling slides, or drafting a report. The agent asks for confirmation before making purchases, so you keep control over any transactions.

Sessions are tied to your account, not your device. That means you can start a task on one machine and pick it up elsewhere without losing context. The integration requires no additional configuration, which is a notable shift from earlier setups that demanded separate tooling.

Security Concerns Remain

Anthropic warns about hidden instructions on websites. The extension remains a potential attack vector for prompt injections, like all LLM interfaces. Malicious pages could attempt to steer the agent's behavior, so the company advises caution.

The #1 Newsletter in AI

Stay ahead of the AI curve

The most important updates, news, and content — delivered weekly.

No spam. Unsubscribe anytime.

Anthropic also recommends against using the extension with banking or health data. Sensitive information, the company says, should not flow through this tool. The warning reflects a broader reality: browser-based agents inherit the risks of the pages they touch.

Availability and Timeline

The update went live as of the article's publication on Aug 13, 2026. It is restricted to paid plans, so free users will not see the new side panel features. The extension is distributed through the Chrome Web Store, where existing users can update and new users can install it.

Before this update, the extension was a lighter tool, useful mainly for reading pages and filling forms. Now it carries the full weight of Cowork's agentic capabilities, bringing automation directly into the browsing experience.

What This Means for Browser Automation

This is the first time skills, plugins, and connectors function inside the browser environment. Previously, such capabilities required separate applications or manual transfers between tools. The shift collapses that distance, letting a single panel handle research, creation, and follow-up actions.

The security trade-offs are real, though. Prompt injection risks persist, and Anthropic's guidance about banking and health data signals limits on where the tool should be used. For everyday productivity tasks, the integration offers a direct path from browsing to finished work.

Related on Neura Market

More from Neura News

AI Models

42 Mathematicians Urge Royal Society to Warn Government and Media About AI Existential Risk

Forty-two mathematical fellows, including Fields Medal winners Martin Hairer, Peter Scholze, and Wendelin Werner, have signed an open letter urging the Royal Society to warn the UK government and media about existential risks from advanced AI. The letter follows recent breakthroughs in which leading models solved open research problems, including a Millennium Problem. None of the signatories are affiliated with AI companies. The group warns that AI labs' estimates of existential risk above ten percent must not be dismissed as hype, and that by the time the situation becomes obvious to the public, it may be too late to act.

Sep 18·2 min read
Developer

Steve Yegge Shuts Down Gas Town After Failing to Build Anything Else With It

Steve Yegge shut down Gas Town, his ultra-vibed coding agent orchestrator, after admitting he never built anything else with it despite heavy subscription spend. Databricks reported a 60% coding spend increase after rolling out GPT-6 Astra to 3,500 engineers, OpenAI published a misalignment disclosure framework with six case reports, and Xiaomi ran MiMo-V2.6 RL training in public with live telemetry.

Sep 18·21 min read