skills/mukul975/anthropic-cybersecurity-skills/exploiting-ms17-010-eternalblue-vulnerability/Gen Agent Trust Hub
exploiting-ms17-010-eternalblue-vulnerability
Warn
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The Python scripts
scripts/agent.pyandscripts/process.pyutilize thesubprocessmodule to execute external binaries likenmap. These scripts pass user-provided network targets directly as command-line arguments. - [REMOTE_CODE_EXECUTION]: The skill contains exhaustive technical workflows for exploiting the MS17-010 vulnerability, including specific Metasploit module configurations (e.g.,
exploit/windows/smb/ms17_010_eternalblue) designed to achieve remote command execution on vulnerable servers. - [EXTERNAL_DOWNLOADS]: The
scripts/process.pyscript requires external Python dependencies, specifically theimpacketandrichlibraries, to facilitate its network communication and report formatting tasks. - [DATA_EXFILTRATION]: The documentation in
references/workflows.mdprovides explicit commands and procedures for extracting sensitive system credentials (e.g., usinghashdumpandload kiwi) once a target system is compromised. - [REMOTE_CODE_EXECUTION]: The skill provides detailed instructions for maintaining long-term access to compromised systems through the use of automated persistence mechanisms within the Meterpreter framework.
- [COMMAND_EXECUTION]: The skill exhibits an attack surface for indirect prompt injection:
- Ingestion points: Network targets and IP addresses provided via CLI arguments and processed by the scanning scripts.
- Boundary markers: None; target data is directly interpolated into command strings.
- Capability inventory: Execution of external binaries through
subprocessand raw SMB protocol negotiation. - Sanitization: Partial; the scripts perform basic IP/CIDR validation but do not fully sanitize target strings before they are passed to system commands.
Audit Metadata