Security audit
Security checks for vulnerabilities and agentic risk
This is a small netcat-style network utility skill with expected but broadly scoped network-use instructions, and no evidence of hidden persistence, credential access, or malicious behavior.
Use this only on hosts and networks you own or are explicitly authorized to test. Before using scan or listen-style commands, confirm the target host, port range, and whether data may be sent or received. Be aware that the bundled script appears limited to a simple TCP connection check despite broader capabilities described in the skill text.
The skill advertises capabilities for port scanning, listen mode, and arbitrary data transfer without any safety guidance, authorization boundaries, or warnings about legal and security implications. In an agent context, this omission can normalize potentially intrusive network behavior and increase the chance of misuse against unauthorized hosts or exposure of sensitive data.
No suspicious patterns detected.
