replay-regression
Warn
Audited by Socket on Oct 20, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill is internally coherent, but it equips the agent to perform live PoC replay and reachability testing against real deployments, which is a high-risk offensive security capability for an AI agent. There is no clear credential theft, exfiltration path, or malicious installer behavior in the provided text.
Confidence: 86%Severity: 72%
Audit Metadata