network-protocol-attacks

MALICIOUS. This skill’s stated purpose is to help an AI agent conduct network attacks, capture and relay credentials, evade detection, and chain into further unauthorized access. The footprint is internally consistent with offensive intrusion, but that purpose itself makes it high-risk and incompatible with safe general agent use.

The provided fragment is a highly explicit offensive playbook for network poisoning and NTLM credential capture/relay leading to Active Directory compromise. While it is not a code library, it clearly describes malicious behaviors (credential harvesting, forced authentication via poisoning, relay to LDAP/HTTP/MSSQL, and post-exploitation secret dumping). This content presents an extremely high security risk if used or distributed within a software supply chain context. Further review of surrounding repository context would be needed, but based solely on this fragment, it is effectively malicious guidance.