Sunday, October 18, 2026

Critical Adobe Acrobat Chrome Extension Flaw “HermeticReader” Lets Hackers Hijack WhatsApp Chats of 300M+ Users

Guardio Labs has disclosed a critical vulnerability chain in the Adobe Acrobat Chrome extension that could allow a malicious website to hijack and exfiltrate rendered WhatsApp Web data from affected users.

This vulnerability is tracked as CVE-2026-48294 and has impacted Adobe Acrobat extension version 26.5.2. The extension is installed across approximately 329 million browsers.

Adobe has reportedly acknowledged the report, developed a fix, and released a patched version of the extension over one weekend following responsible disclosure.

Critical Adobe Acrobat Chrome Extension Flaw “HermeticReader”

The issue, dubbed “HermeticReader” by researchers, did not require malware, phishing credentials, an Adobe account, a WhatsApp vulnerability, or theft of session cookies.

Instead, an attacker could host a malicious static webpage and rely on a victim to simply visit it while signed into WhatsApp Web and using a vulnerable version of the Adobe Acrobat extension.

Attack Chain Flow (Source: Guard Labs)
Attack Chain Flow (Source: Guard Labs)

This vulnerability exploited Acrobat’s newly introduced WhatsApp Web integration, internally known as Hermes, which was designed to support PDF workflows within the messaging platform.

Guardio Labs discovered that several weak trust boundaries could be exploited to turn a webpage into a cross-origin mechanism for exfiltrating WhatsApp data.

First, the extension exposed web-accessible pages that accepted JSON data through URL query parameters. These pages operated with the extension’s trusted ‘chrome-extension://’ origin and relayed attacker-controlled fields to the extension service worker via chrome.runtime.sendMessage. Although one control field was overwritten during processing, much of the message content remained within the attacker’s control.

The service worker reportedly failed to validate the sender of inbound messages. This allowed untrusted websites to invoke storage operations, including arbitrary writes to the extension’s local storage.

Researchers used this capability to alter a developer feature-flag override value, enabling the dormant Hermes integration engine. Specifically, writing the `dc-cv-hermes` value into the `floodgate-add` feature-flag key activated WhatsApp-related functionality without any user interaction.

The next step required identifying the target WhatsApp Web tab. Guardio Labs found a weak hostname check that trusted domains beginning with strings like “www.google..”.

An attacker-controlled hostname resembling `www.google.attacker-domain.com` could satisfy this condition and trigger an Acrobat overlay flow.

This flow inadvertently exposed the attacker page’s Chrome tab ID through an injected extension iframe. Since Chrome tab IDs are allocated sequentially, the attacker could open WhatsApp Web in a new tab and predict its ID.

With the expected WhatsApp Web tab ID, attackers could send Hermes commands to Acrobat’s content script inside the WhatsApp page. The content script accepted a broad range of privileged DOM operations without authenticating the sender.

These commands included injecting HTML, manipulating arbitrary DOM elements, invoking methods on the window, accessing storage values, and relaying messages back to the extension service worker.

Although WhatsApp’s content security policy prevented straightforward script injection, researchers demonstrated a more subtle path for exfiltration.

The exploit injected an attacker-controlled form into WhatsApp Web, manipulated DOM nodes to move the live WhatsApp page body into a form option element, and submitted the form to an attacker-controlled server.

Because an option element without a value attribute submits its text content, the resulting POST request could include rendered WhatsApp Web text, such as chat lists, contact names, message previews, profile details, and visible conversations.

This vulnerability highlights the risks associated with browser extensions that bridge websites, privileged extension APIs, and third-party messaging platforms.

Guardio Labs noted that its AI-assisted research workflow expedited the discovery process by unpacking and analyzing hundreds of obfuscated JavaScript files, mapping service worker message flows, and identifying reachable attack paths in hours rather than weeks.

Users are urged to ensure that the Adobe Acrobat Chrome extension is updated to the latest version. Organizations should also review their browser extension inventories, prioritize rapid patching for extensions with broad site permissions, and consider limiting unmanaged extensions in enterprise environments.

ALERT: 20+ government sites delivered malware to businesses and citizens. See full attack research to check your own exposure.

Divya
Divya
Divya is a Senior Journalist at GBhackers covering Cyber Attacks, Threats, Breaches, Vulnerabilities and other happenings in the cyber world.

Hot this week

How To Access Dark Web Anonymously and know its Secretive and Mysterious Activities

What is Deep Web The deep web, invisible web, or...

How to Build and Run a Security Operations Center (SOC Guide) – 2023

Today’s Cyber security operations center (CSOC) should have everything...

Russian Hackers Bypass EDR to Deliver a Weaponized TeamViewer Component

TeamViewer's popularity and remote access capabilities make it an...

Web Server Penetration Testing Checklist – 2026

Web server pentesting is performed under three significant categories: identity,...

ATM Penetration Testing – Advanced Testing Methods to Find The Vulnerabilities

ATM Penetration testing, Hackers have found different approaches to...

Hackers Use ConnectWise SkyKick Credentials to Access Microsoft 365, Halting Cloud Backups

Attackers abused compromised ConnectWise Cloud Backup credentials to access...

Anthropic’s Claude AI Uncovers Origins of NSA-Provided Cryptographic Keys After 25 Years

Anthropic says an internal Claude research model reconstructed how...

OpenBSD Kernel Flaw Dating Back 37 Years Enables Local Privilege Escalation to Root

A newly disclosed OpenBSD kernel exploitation chain combines a...

GhostAction Hackers Hide Credential-Stealing Malware in Fake Font Files Across 94 GitHub Repositories

GhostAction attackers have expanded their GitHub credential theft campaign...

SideCopy Hackers Target Indian Organizations With ReverseRAT Malware to Steal Data

Indian authorities have warned that SideCopy hackers are targeting...

New GhostBat Android Malware Targets 60 Banking and UPI Apps to Steal PINs

A new GhostBat RAT variant targets 60 Android financial...

New ROMULUS Malware Platform Uses Six Servers to Deliver Multiple Malicious Payloads

ROMULUS, a shared ClickFix delivery platform operating across six...

Related Articles

Recent News