Browser-Agents

A browser window melting away while a code terminal glows brightly in the foreground, set against a dark blue gradient background

Google Abandons Browser Agents for Coding Agents — Project Mariner Team Shakeup Signals Industry Shift

If you want to understand where the AI industry thinks agentic AI is heading, watch where Google moves its engineers. This week, Wired reported that Google is restructuring Project Mariner — its Chrome browser AI agent, which lets AI systems browse the web and click through interfaces the way a human would. Engineers from the Mariner team are being pulled to work on higher-priority coding agent projects. A Google spokesperson confirmed the changes, telling Wired that Mariner’s computer-use capabilities will live on inside the company’s broader agent strategy — but the dedicated browser agent team as it existed is being reorganized. ...

A Chrome browser window with glowing MCP protocol connection lines extending outward to abstract AI agent nodes

Chrome 146 Ships Native MCP Support — AI Agents Can Now Access Live Browser Sessions

When Chrome ships a feature, it ships to roughly 3.4 billion browsers simultaneously. That’s what makes Chrome 146’s native Model Context Protocol (MCP) support such a significant — and potentially consequential — development for the agentic AI ecosystem. What WebMCP Actually Is MCP, for those who need the refresher: it’s Anthropic’s open protocol for connecting AI models to external tools and data sources in a standardized way. The “Web” prefix in WebMCP specifically means browser sessions — live, authenticated, cookie-bearing browser sessions. ...

Abstract dark web of tangled red lines converging on a single bright node, representing hidden manipulation of a connected system

Hackers Are Poisoning Websites to Hijack AI Agents via Indirect Prompt Injection

The attack is elegant in a disturbing way. An adversary doesn’t need to breach your AI infrastructure, compromise your API keys, or exploit a software vulnerability. They just need to get your AI agent to read a web page they control — and then they’re driving. Indirect Prompt Injection (IDPI) is the attack technique where malicious instructions are embedded in content that an AI agent processes: web pages, documents, calendar entries, emails. When the agent reads that content, it encounters instructions that override or subvert its intended behavior. The content tells the agent what to do, and the agent, trained to follow instructions, complies. ...

How to Harden Your OpenClaw Agents Against Indirect Prompt Injection

Indirect Prompt Injection (IDPI) is now confirmed in-the-wild by Palo Alto Unit 42. Adversaries are embedding hidden instructions in web pages and documents to hijack AI agents — and OpenClaw’s browser and research agents are high-value targets. This guide walks through concrete hardening steps you can apply to your OpenClaw deployments today. Prerequisites OpenClaw installed and configured (any recent version) At least one agent with web browsing or document processing capability Basic familiarity with OpenClaw’s skill and session configuration Step 1: Audit Your Agent Attack Surface Before hardening anything, map your exposure. For each agent you run: ...

WebMCP in Chrome 146: Google and Microsoft's Proposed Standard Exposes Structured Website Actions to AI Agents

Something genuinely important is shipping in Chrome 146: an early preview of WebMCP, a W3C draft standard jointly developed by Google and Microsoft that fundamentally changes how AI agents interact with websites. Right now, AI agents that browse the web do so by scraping DOM elements — reading HTML, finding buttons, inferring what actions are available. It’s brittle. A website redesign breaks the agent. A modal renders differently across browsers and the agent gets stuck. This approach works well enough for demos but fails at production scale. ...