exploiting-insecure-deserialization

SUSPICIOUS/HIGH-RISK skill. Its instructions are internally consistent with an offensive penetration-testing purpose, but that purpose gives the agent explicit exploit and remote-command-execution capability, routes data through OAST/interactsh callback infrastructure, and enables real-world attacks against external targets. The GitHub tool sources appear official, so this is not confirmed malware, but it is a dangerous exploit skill and should be treated as high risk.

This is a security assessment utility rather than apparent package malware. Its normal execution scans a user-selected URL and writes findings locally without deserializing input. However, it contains explicitly dangerous exploit-test routines, especially the Python pickle probe that can cause a vulnerable target to execute nslookup, and it disables TLS verification for all requests. The exploit functions are currently unreachable from the provided main path, but they should be treated as dual-use offensive testing code and restricted to authorized targets.