Close Menu
    Facebook X (Twitter) Instagram
    Wifi PortalWifi Portal
    • Blogging
    • SEO & Digital Marketing
    • WiFi / Internet & Networking
    • Cybersecurity
    • Tech Tools & Mobile / Apps
    • Privacy & Online Earning
    Facebook X (Twitter) Instagram
    Wifi PortalWifi Portal
    Home»Cybersecurity»The vulnerability that turns your AI agent against you
    Cybersecurity

    The vulnerability that turns your AI agent against you

    adminBy adminMarch 4, 2026No Comments3 Mins Read
    Facebook Twitter LinkedIn Telegram Pinterest Tumblr Reddit WhatsApp Email
    The vulnerability that turns your AI agent against you
    Share
    Facebook Twitter LinkedIn Pinterest Email

    Zenity Labs disclosed PleaseFix, a family of critical vulnerabilities affecting agentic browsers, including Perplexity Comet, that allow attackers to hijack AI agents, access local files, and steal credentials within authenticated user sessions. The vulnerabilities can be triggered through malicious content embedded in routine workflows, enabling unauthorized actions without user awareness.

    agentic browser vulnerability

    The disclosure includes PerplexedBrowser, a subfamily of vulnerabilities in the Perplexity Comet browser that consists of two distinct exploit paths. Both stem from indirect prompt injection techniques but produce materially different outcomes.

    The first enables zero-click agent compromise that grants access to the local file system and allows data exfiltration while the agent continues returning expected results to the user. The second abuses agent-authorized workflows to manipulate password manager interactions, resulting in credential theft or full account takeover without directly exploiting the password manager itself.

    Agentic browsers represent a new computing model. Unlike traditional browsers that primarily display content, agentic systems interpret instructions, retain authenticated context and autonomously execute actions across applications and services.

    PleaseFix demonstrates how this expanded capability introduces new security risks by extending user trust into automated workflows, exposing sensitive data, credentials and connected systems in ways existing browser and endpoint controls were not designed to detect.

    Zenity Labs’ discovery

    Researchers identified vulnerabilities that allow AI agents to operate autonomously within authenticated browser sessions. When an agent is asked to perform a routine task such as accepting a calendar invite, it can execute actions without human validation and inherit access to data, tools, and workflows the user has authorized.

    PleaseFix represents the evolution of ClickFix, a social engineering technique in which attackers trick users into executing malicious actions. In this case, the technique is applied to AI agents, allowing malicious actions to be triggered without human involvement.

    “This is an inherent vulnerability in agentic systems,” said Michael Bargury, CTO of Zenity. “Attackers can push untrusted data into AI browsers and hijack the agent itself, inheriting whatever access it has been granted. This is an agent trust failure that exposes data, credentials and workflows in ways existing security controls were never designed to see.”

    Exploit 1

    In the first exploit, attacker-controlled content, such as a calendar invite, triggers autonomous execution in the Perplexity Comet browser when a user asks the agent to perform a routine task (a 0-click vulnerability).

    No prompts or user interaction are required. The agent autonomously accesses the local file system and exfiltrates the contents to an attacker-controlled endpoint, while still returning the expected response to the user.

    Exploit 2

    The second exploit also begins with an attacker-controlled trigger, allowing the attacker to assume agent privileges and abuse agent-authorized workflows that allow access to password management tools. Without exploiting password managers directly, attackers can manipulate agent task execution to steal individual stored credentials or take over the user’s account. These actions occur within a legitimate, authenticated session.

    Zenity Labs responsibly disclosed the PleaseFix vulnerability and exploits. Perplexity addressed the underlying browser-side agent execution issue prior to public disclosure.

    Webinar: The True State of Security 2026

    Agent Turns vulnerability
    Share. Facebook Twitter Pinterest LinkedIn Tumblr Telegram Email
    Previous ArticleSeraphinite Accelerator WordPress Plugin Vulnerabilities Affect 60K Sites
    Next Article The Galaxy S26 Ultra makes it clear this feature isn’t coming back
    admin
    • Website

    Related Posts

    Fake Laravel Packages on Packagist Deploy RAT on Windows, macOS, and Linux

    March 4, 2026

    Paint maker giant AkzoNobel confirms cyberattack on U.S. site

    March 4, 2026

    VMware Aria Operations Vulnerability Exploited in the Wild

    March 4, 2026
    Add A Comment
    Leave A Reply Cancel Reply

    Search Blog
    About
    About

    At WifiPortal.tech, we share simple, easy-to-follow guides on cybersecurity, online privacy, and digital opportunities. Our goal is to help everyday users browse safely, protect personal data, and explore smart ways to earn online. Whether you’re new to the digital world or looking to strengthen your online knowledge, our content is here to keep you informed and secure.

    Trending Blogs

    I have tried dozens of cloud storage apps, but I keep coming back to OneDrive due to this privacy feature

    March 4, 2026

    Fake Laravel Packages on Packagist Deploy RAT on Windows, macOS, and Linux

    March 4, 2026

    How to Focus on Topics (Not Keywords) in Your SEO Strategy

    March 4, 2026

    The Galaxy S26 Ultra makes it clear this feature isn’t coming back

    March 4, 2026
    Categories
    • Blogging (32)
    • Cybersecurity (595)
    • Privacy & Online Earning (88)
    • SEO & Digital Marketing (374)
    • Tech Tools & Mobile / Apps (731)
    • WiFi / Internet & Networking (106)

    Subscribe to Updates

    Stay updated with the latest tips on cybersecurity, online privacy, and digital opportunities straight to your inbox.

    WifiPortal.tech is a blogging platform focused on cybersecurity, online privacy, and digital opportunities. We share easy-to-follow guides, tips, and resources to help you stay safe online and explore new ways of working in the digital world.

    Our Picks

    I have tried dozens of cloud storage apps, but I keep coming back to OneDrive due to this privacy feature

    March 4, 2026

    Fake Laravel Packages on Packagist Deploy RAT on Windows, macOS, and Linux

    March 4, 2026

    How to Focus on Topics (Not Keywords) in Your SEO Strategy

    March 4, 2026
    Most Popular
    • I have tried dozens of cloud storage apps, but I keep coming back to OneDrive due to this privacy feature
    • Fake Laravel Packages on Packagist Deploy RAT on Windows, macOS, and Linux
    • How to Focus on Topics (Not Keywords) in Your SEO Strategy
    • The Galaxy S26 Ultra makes it clear this feature isn’t coming back
    • The vulnerability that turns your AI agent against you
    • Seraphinite Accelerator WordPress Plugin Vulnerabilities Affect 60K Sites
    • Oukitel WP63 is a phone I would stock for the apocalypse
    • Paint maker giant AkzoNobel confirms cyberattack on U.S. site
    © 2026 WifiPortal.tech. Designed by WifiPortal.tech.
    • Home
    • About Us
    • Contact Us
    • Privacy Policy
    • Terms and Conditions
    • Disclaimer

    Type above and press Enter to search. Press Esc to cancel.