Chrome Zero-Days Keep Coming, AppGuard Keeps Stopping Them

On February 17, 2026, CISA added CVE-2026-2441 — a critical use-after-free in Google Chrome’s CSS component — to the Known Exploited Vulnerabilities (KEV) catalog. This marks the first confirmed Chrome zero-day actively used in attacks this year, enabling remote code execution via crafted HTML pages.

Chrome zero-days in the wild are not rare. They follow a clear pattern.

Chrome’s First KEV by Year

(This table highlights the date when the first Chrome vulnerability was added to the KEV catalog each year, showing the recurring early-year threat pattern.)

Year Date Added CVE ID
2021 November 3 Multiple
2022 January 10 CVE-2020-6572
2023 April 17 CVE-2023-2033
2024 January 2 CVE-2023-7024
2025 March 27 CVE-2025-2783
2026 February 17 CVE-2026-2441

While the first of each year stands out, Chrome has far more exploited vulnerabilities overall.

AppGuard Controls Mitigation for All Chrome KEVs (2021–2026)

(This table summarizes how AppGuard’s three controls mitigate post-exploit actions across every Chrome KEV in the period.)

Year Number of KEVs Launch Contain Isolate
2021 18 Yes Yes Yes
2022 13 Yes Yes Yes
2023 6 Yes Yes Yes
2024 4 Yes Yes Yes
2025 7 Yes Yes Yes
2026 1 Yes Yes Yes
  • Launch: Restricts executions from high-risk user-space folders.
  • Contain: Limits the entire Chrome process tree, including default-deny of unauthorized child processes (with full inheritance to any spawned children).
  • Isolate: Protects registry keys, files, memory, and credential stores from unauthorized access or modification.

By enforcing these controls, AppGuard also significantly reduces the overall attack surface, slashing the potential harm from this never-ending stream of Chrome vulnerabilities.

The Top 20 Most Common Post-Exploit Actions After Browser RCE That AppGuard Controls Stop

(Mandiant’s M-Trends 2025 and MITRE ATT&CK data show these are the actions attackers take immediately after gaining code execution inside Chrome.)

Rank Post-Exploit Action (Human Terms) MITRE ATT&CK Primary Actor Process AppGuard Control(s)
1 Launches PowerShell or cmd.exe as child T1059.001 chrome.exe spawns child Chrome process tree containment blocks malicious actions
2 Writes to registry Run key for persistence T1547.001 chrome.exe or child Chrome process tree containment blocks malicious actions; Isolation of objects blocks malicious actions
3 Launches child for reconnaissance commands T1059.001 chrome.exe spawns child Chrome process tree containment blocks malicious actions
4 Drops executable to %TEMP% or Downloads then launches it T1105 chrome.exe Launches from High-risk folders blocked; Chrome process tree containment blocks malicious actions
5 Spawns wscript/cscript for malicious script T1059.007 chrome.exe spawns child Chrome process tree containment blocks malicious actions
6 Loads malicious DLL via side-loading T1574.002 chrome.exe Launches from High-risk folders blocked
7 Dumps LSASS memory for credentials T1003.001 chrome.exe or child Isolation of objects blocks malicious actions
8 Uses COM objects to access files or registry T1559 chrome.exe Chrome process tree containment blocks malicious actions; Isolation of objects blocks malicious actions
9 Creates scheduled task for persistence T1053.005 chrome.exe or child Chrome process tree containment blocks malicious actions; Isolation of objects blocks malicious actions
10 Attempts to read Chrome’s saved passwords or cookies T1555.003 chrome.exe Isolation of objects blocks malicious actions
11 Writes malicious script to user-space then launches it T1105 chrome.exe Launches from High-risk folders blocked; Chrome process tree containment blocks malicious actions
12 Uses rundll32 proxy execution T1218 chrome.exe spawns child Chrome process tree containment blocks malicious actions
13 Creates new Windows service for persistence T1543.003 chrome.exe or child Chrome process tree containment blocks malicious actions; Isolation of objects blocks malicious actions
14 Writes to Startup folder for persistence T1547.001 chrome.exe or child Launches from High-risk folders blocked; Chrome process tree containment blocks malicious actions
15 Accesses protected files outside the sandbox T1083 chrome.exe Isolation of objects blocks malicious actions
16 Uses PowerShell child for malicious commands T1059.001 chrome.exe spawns child Chrome process tree containment blocks malicious actions
17 Modifies registry for privilege escalation T1548 chrome.exe or child Chrome process tree containment blocks malicious actions; Isolation of objects blocks malicious actions
18 Performs fileless execution in memory T1055 chrome.exe Chrome process tree containment blocks malicious actions
19 Uses regsvr32 to load malicious DLL T1218 chrome.exe spawns child Chrome process tree containment blocks malicious actions
20 Attempts to inject code into system processes T1055 chrome.exe Chrome process tree containment blocks malicious actions

With memory-based attacks and process injection techniques on the rise, AppGuard’s Contain and Isolate controls are increasingly critical. They prevent compromised browsers from stealing credentials from memory or injecting code into other processes.

Like all first exploits in the wild of Chrome, AppGuard mitigates the risk, not by pattern-matching guessing like EDR/XDR but with controls-based endpoint protection that blocks malicious activities—before damage occurs.

Detection tools are necessary but not sufficient. Add AppGuard’s controls-based layer for practical, low-friction protection that complements what you already have.

Ready to stop what detection misses? Get more info now.

Subscribe

Subscribe to our blog to receive email notifications when new posts are added!

Loading