The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent on how to use msfconsole and related command-line utilities to perform network enumeration and security testing.\n- [DYNAMIC_EXECUTION]: The documentation includes workflows for generating standalone executable payloads using msfvenom for various target platforms, including Windows, Linux, and Android.\n- [PRIVILEGE_ESCALATION]: The skill describes the use of Metasploit modules, such as getsystem and bypassuac, which are designed to elevate user privileges on compromised systems during authorized testing.\n- [PERSISTENCE]: Instructions are provided for utilizing post-exploitation modules like persistence_exe and sshkey_persistence to maintain access to targets across sessions.\n- [CREDENTIALS_UNSAFE]: The documentation details methods for harvesting sensitive information, including dumping password hashes (hashdump) and collecting credentials from target memory or files.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface where the agent processes output from untrusted external targets. This is addressed by a mandatory confirmation gate that requires explicit user authorization before performing any active or intrusive operations.

metasploit-framework