[!WARNING] Final verdict available at the end of this report.
This JAR scan reviewed all 35 classes in the file and successfully parsed and decompiled all of them. It found 7 classes with security-relevant behavior, representing 20% of the JAR’s classes. The detected indicators include local file access, outbound HTTP connections, a process-launching API, and cryptographic hashing.
Most findings are concentrated in the com.payfake application code, particularly configuration and license-validation classes. The most important areas for closer inspection are the hardware-ID collection code, the use of ProcessBuilder, and connections to three licensing-related domains.
Some scan signals can be normal for software with settings, license activation, or online validation. The class-level evidence below helps distinguish ordinary configuration and licensing functions from behavior that would be more strongly associated with malware.
No high-signal malicious behavior was detected.
The scan did not identify evidence of credential theft, browser/token grabbing, hidden payload downloads, persistence mechanisms, destructive file wiping, remote shell behavior, or obvious data exfiltration. It did find that the JAR can read and manage local configuration files, collect a Windows-related machine identifier, hash it with SHA-256, and send licensing-related requests to external services.
The presence of a process-launching API deserves review, but the only flagged method is named windowsMachineGuid(), which is consistent with obtaining a machine-specific hardware identifier for license locking. The report does not show arbitrary commands being launched or commands received from a remote server.
The evidence is not consistent with a clear remote-access trojan (RAT).
A RAT normally includes remote command handling, hidden remote control, persistence, screen/input access, file theft, or command-and-control traffic. This report shows outbound licensing traffic and a hardware-ID function, but no evidence of remote command execution, a backdoor listener, persistence, or remote-control features.
All seven flagged classes are in the JAR’s com.payfake namespace, so these signals appear to come from the application’s own code rather than a known shaded third-party dependency.
| Class | Severity | Behavior | Evidence |
|---|---|---|---|
com.payfake.Configs |
Low | Reads, saves, duplicates, deletes, and lists configuration files | Uses Java NIO file APIs in configuration-management methods such as save, load, and delete |
com.payfake.Settings |
Low | Loads and saves settings | Uses Java NIO file APIs |
com.payfake.license.Hwid |
Medium | Creates a Windows machine identifier and hashes it | windowsMachineGuid() uses ProcessBuilder; sha256() uses cryptographic hashing |
com.payfake.license.KeyAuth |
Medium | Sends license requests to KeyAuth | HTTPS POST request to https://keyauth.win/api/1.3/ |
com.payfake.license.Keygen |
Medium | Validates and activates licenses tied to a machine ID | Requests to https://api.keygen.sh/v1/accounts/ and Java HTTP client use |
com.payfake.license.License |
Medium | Contains a custom licensing-service address | Hardcoded URL: https://payfake-licenses.noahmenkyna667.workers.dev |
com.payfake.license.LicenseClient |
Medium | Sends license validation requests | Java HTTP client in validate(String baseUrl, String key, String hwid) |
No flagged classes were identified as recognizable bundled third-party libraries such as bStats, HSQLDB, SLF4J, Apache Commons, or HikariCP. The scan’s suspicious findings are instead located in the JAR’s own com.payfake code.
The JAR contains three external HTTPS destinations:
https://keyauth.win/api/1.3/
Used by com.payfake.license.KeyAuth to send a POST request. KeyAuth is a license/authentication platform, so this is consistent with online license checking.
https://api.keygen.sh/v1/accounts/
Used by com.payfake.license.Keygen for key validation and machine activation. The method signatures explicitly include a license key and HWID (hardware identifier).
https://payfake-licenses.noahmenkyna667.workers.dev
A custom licensing endpoint hosted through Cloudflare Workers. The report confirms that the URL is embedded in the JAR, but does not show the request body, when it is contacted, or exactly what data is sent.
No IP addresses, download URLs, webhook endpoints, or known command-and-control endpoints were identified in the report.
The scanner had strong coverage of this file:
There are no reported signs of code that could not be read, decompiler-breaking tricks, or obvious obfuscation-related anomalies. The JAR targets Java class version 65, which corresponds to Java 21.
The scanner flagged SHA-256 cryptographic code as requiring review, but a SHA-256 helper is normal when deriving or comparing a machine-bound license identifier. It is not, by itself, evidence of encryption-based concealment or malware.
Several findings have ordinary software explanations:
Configs and Settings classes appear designed to save, load, copy, check, and delete named settings profiles.There are no major malware red flags such as token theft, hidden downloads, remote-control functions, destructive commands, or persistence.
However, there are unresolved concerns:
ProcessBuilder while gathering a Windows machine identifier. The method name suggests a licensing purpose, but process-launching APIs are powerful and should be treated carefully.payfake-licenses.noahmenkyna667.workers.dev. The scan cannot confirm what information is exchanged with that endpoint.8a1b3ec69a9926e4a01e6187bf20fe3fad3ad4305cd1e313c5bb5039146575cfStatic JAR scanning reduces risk by examining the code and embedded strings, but it cannot prove a file is completely safe. A scan may not reveal behavior delivered later by a remote server, behavior triggered only under certain conditions, or data sent inside encrypted HTTPS requests.
Yes. A JAR is Java program code and can perform harmful actions when run if it has the necessary permissions. This report did not find strong malware-style behavior, but that does not make the file automatically virus-free.
Check its source, compare its published hash, inspect network destinations, review process/file-access behavior, and scan it with more than one security tool. For this JAR, the main items to verify are its licensing domains and the ProcessBuilder use in its HWID code.
It can identify many RAT-like warning signs, including process execution, suspicious remote connections, persistence, hidden downloads, and remote command behavior. This report does not show the combination of indicators normally expected from a Minecraft RAT.
No. Some developers obfuscate code to protect it from copying or tampering. In this case, the scan reported no decompiler failures or anomaly indicators, so there is no evidence of significant anti-analysis obfuscation.
Verdict: Use caution
Risk level: Moderate
Confidence: Moderate
Should I trust this JAR, and why? Use caution with this JAR because its code is fully readable and does not show clear credential theft, RAT behavior, hidden downloads, or destructive actions, but it does collect a machine identifier and contact several licensing services. The ProcessBuilder use appears connected to Windows HWID collection rather than arbitrary command execution, yet it remains a powerful API in application-owned code. The custom Cloudflare Workers license domain is not explained by the report, so its data handling cannot be verified from static analysis alone. The “Cracked” filename adds supply-chain uncertainty, especially if the file was not obtained from the original developer.