The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides fully functional webshell payloads for multiple platforms (JSP, PHP, ASPX) and Java memory-resident shells (Filter, Servlet, Listener) designed for arbitrary command execution on remote servers.
[COMMAND_EXECUTION]: Instructions guide the user or agent to create local exploit files and use curl to perform offensive actions, such as HTTP PUT exploitation and multi-part form upload attacks.
[DATA_EXFILTRATION]: Specifically demonstrates how to read highly sensitive system files like /etc/shadow and /etc/passwd and provides a reverse shell payload to establish an outbound command-and-control connection.
[REMOTE_CODE_EXECUTION]: Details exploitation methods for known vulnerabilities like Tomcat CVE-2017-12615, including bypass techniques using trailing slashes, encoded spaces, and NTFS data streams to upload executable files.
[PROMPT_INJECTION]: Outlines a methodology for log injection via the User-Agent header to achieve remote code execution through Local File Inclusion (LFI) vulnerabilities.
[REMOTE_CODE_EXECUTION]: Provides multiple obfuscated PHP payloads (using XOR encryption, Base64 encoding, and callback functions) specifically intended to evade detection by Web Application Firewalls (WAFs) and Antivirus (AV) software.

webshell-deploy