avatar

Kaung Khant Ko

@kaungkhantko
Trusted contributorCompletionist
13 followers1,365 downloads1,865 copies115 likes21,259 views

410 detections

Identifies the execution of PowerShell with suspicious base64 decoding activity.
This behavior is often observed during malware installation leveraging PowerShell.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
7 months ago
1082
mshta.exe is actively abused by DeepLoad early in the attack chain to reach attacker staging infrastructure.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
7 months ago
150126
Detects the creation of specific disguised shortcut (.lnk) files on drives, indicative of DeepLoad's USB propagation mechanism. DeepLoad spreads to adjacent endpoints by dropping over 40 disguised .lnk files onto connected USB drives.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
7 months ago
50100
Detects PowerShell unexpectedly spawning specific built-in Windows processes (LockAppHost.exe, makecab.exe, Magnify.exe). DeepLoad uses Add-Type compiled C# to inject into these processes to hide its fileless payload via asynchronous procedure call (APC) injection.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
7 months ago
3087
Detects the creation or execution of 'filemanager.exe' in the user's Downloads folder. DeepLoad uses WMI subscriptions that survive standard cleanup to drop this specific credential stealer into the Downloads folder three days after initial infection.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
7 months ago
10071
Detects OAuth authorization requests to Microsoft or Google that use the 'prompt=none' parameter in combination with an intentionally invalid or missing 'scope'. This combination is used to force a silent error redirect to an attacker-controlled URI, bypassing some security filters.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
8 months ago
190385
Detects a file download where the referrer URL is a Microsoft Entra ID OAuth endpoint returning an error. This pattern is indicative of an attacker abusing OAuth's error-handling redirection mechanism to lead a user to a malicious payload on an attacker-controlled domain.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
8 months ago
160146
Detects the installation of a browser extension that requests both the 'declarativeNetRequest' permission and host permissions for 'gemini.google.com'. This specific combination of permissions was the vector for CVE-2026-0628, a vulnerability that allowed an extension to inject code into the privileged Gemini side panel, leading to privilege escalation and unauthorized access to local files, camera, and microphone.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
8 months ago
150142
Detects when the Google Chrome process (chrome.exe) reads or creates an unusually high number of files in sensitive user directories within a short time frame. This behavior is indicative of file and directory discovery (T1083) and could be a post-exploitation step of a browser-based attack, such as the Gemini panel hijacking described in CVE-2026-0628, where a compromised browser process is used to access local files outside of its sandbox.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
8 months ago
15090
Detects the legitimate 'steam_monitor.exe' loading 'crashhandler.dll' from a non-standard directory like ProgramData. This side-loading technique was observed in campaigns delivering Korplug/GreedyRobin malware, often following initial access via OAuth redirection abuse.
avatar
Kaung Khant Ko@kaungkhantko
avatar
Detections.ai Community
8 months ago
9073
Page 35 of 41