Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent home-network scanner, but it needs Review because it can actively scan local or public targets without strong scope checks or clear warnings.
Install only if you are comfortable running active nmap scans from this machine and will use it only on networks and public IPs you own or are authorized to test. Review the scanner before use, avoid deep scans by default, and consider adding target validation, explicit confirmation, an allowlist, and disclosure or opt-out for the public-IP lookup.
scanner.py:43Unvalidated Scan Target Permits Nmap Option Injection
The skill exposes shell and network-capable operations but does not declare any explicit tool scope or permission boundaries. That increases the chance an agent platform will run active scanning behavior without clear user consent, policy gating, or least-privilege enforcement, which is risky for a network-auditing skill.
The skill invites active LAN discovery and external-facing scans, including deep vulnerability scanning of a router, but does not warn users that these actions may disrupt fragile devices, trigger IDS/ISP abuse controls, or violate acceptable-use policies. In a home-network context, the behavior is expected functionally, but the missing warning still makes misuse and unintended harm more likely.
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
Ensure nmap is installed on the host system:
sudo apt install nmap
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
def get_network_topology():
try:
# Extract default gateway using ip route
route_out = subprocess.run(['ip', 'route'], capture_output=True, text=True).stdout
gateway_match = re.search(r'default via ([0-9\.]+)', route_out)
gateway = gateway_match.group(1) if gateway_match else "Unknown"
The skill contacts an external service to determine the public IP without any disclosure, consent, or privacy notice. This leaks network-identifying metadata to a third party and may violate user expectations or organizational policy, especially in a security-auditing tool operating on local infrastructure.
The code performs an active LAN sweep with nmap automatically and without any user-facing warning or confirmation. Active scanning can trigger IDS/IPS alerts, disrupt sensitive devices, and violate acceptable-use rules if run in unmanaged or third-party environments.
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
subnet = f"{parts[0]}.{parts[1]}.{parts[2]}.0/24"
# Run nmap ping sweep
result = subprocess.run(['nmap', '-sn', subnet], capture_output=True, text=True)
hosts = []
for line in result.stdout.split('\n'):
The skill allows port and service scanning of user-supplied targets with no warning, authorization check, or restriction on scope. In context, this is more dangerous because the skill is explicitly a network security agent, making it a ready-made reconnaissance capability that can be repurposed against arbitrary internal or external targets from the host running it.
The code passes a user-supplied target directly into nmap, enabling arbitrary host scanning from the environment where this skill runs. Although shell injection is mitigated by argument-list invocation, this still creates a scanning primitive that can be abused for unauthorized reconnaissance against internal or external systems.
else:
cmd = ['nmap', '-F', '-sV', ip_address]
result = subprocess.run(cmd, capture_output=True, text=True)
open_ports = []
for line in result.stdout.split('\n'):
No suspicious patterns detected.