social-engineering
Audited by Socket on Sep 19, 2026
6 alerts found:
Malwarex3Securityx3MALICIOUS. This skill’s core function is social-engineering attack execution: phishing, credential harvesting, cookie theft, impersonation, and defense evasion. The footprint is internally consistent with that offensive purpose, but that purpose itself enables direct theft and real-world harm, making the skill high risk and clearly unsafe.
The fragment contains clear malicious and offensive capabilities, including credential theft, phishing, session-token capture, remote payload execution, credential validation, badge cloning, and physical-access bypass. Authorization disclaimers do not neutralize the direct implementation of these capabilities. It should not be included in or executed by a software supply-chain package.
This material intentionally implements a credential-phishing campaign: it sends deceptive password-reset emails, presents a spoofed SSO login, collects submitted usernames and passwords, tracks recipient activity, and reports harvested credentials. Although framed for authorized assessments, the functionality is directly usable for credential theft. The hardcoded secret placeholders and verify=False settings add operational security weaknesses. It should not be deployed against real users or systems without strict authorization, scope controls, and a design that avoids collecting real passwords.
This is documentation for dual-use security tools, not executable malware. It contains substantial high-risk guidance for phishing, credential harvesting, reconnaissance, brute forcing, and evasion of detection. The examples should only be used in explicitly authorized assessments, with TLS enabled, administrative interfaces restricted, default credentials changed, and secrets kept out of command-line arguments. No direct package backdoor, exfiltration implementation, or malicious persistence mechanism is present in the supplied material.
This is a social-engineering and offensive-security test-plan document rather than malicious executable package code. It contains explicit procedures for phishing, credential harvesting, MFA interception, reverse-shell delivery, spoofing, and physical-security attacks. The artifact itself shows no malware or package-level persistence behavior, but its instructions could enable real credential theft and unauthorized access if misused. Use only with documented authorization, synthetic credentials, isolated infrastructure, and non-deployable payloads.
The fragment appears to be social-engineering and red-team training material, but it contains operational badge cloning, USB autorun, hidden PowerShell execution-policy bypass, remote script download, and possible credential-harvesting instructions. The authorization disclaimers reduce evidence of malicious intent but do not make the payloads safe. The remote gather.ps1 behavior is unknown and requires separate review. Treat the fragment as high-risk dual-use content and do not execute the payloads outside a tightly controlled, explicitly authorized test environment.