cache-poisoning

SUSPICIOUS: the skill is internally consistent as a cache-poisoning hunting guide, but its core function is offensive security testing against live targets, including authenticated cache-deception scenarios and DoS-style poisoning checks. No clear malicious exfiltration or hidden behavior is present, yet the exploit-focused capability makes it high risk for agent use. Install trust is mostly acceptable except for the WCVS pip example, which is inconsistent with official docs and should be treated cautiously.