hunt-cache-poison

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is explicit, actionable attacker guidance for cache poisoning and web-cache deception—including header injection, request smuggling, and payloads to achieve mass XSS, account takeover, affiliate fraud, and DoS—indicating clear malicious intent and high abuse risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The runtime workflow depicted in SKILL.md is an attacker “hunting methodology” for cache poisoning, which requires sending crafted HTTP requests (including attacker-controlled headers/parameters) to a target and analyzing the HTTP responses; it does not describe an agent ingesting outsider-authored free text from an email/chat/inbox/feed/issue tracker without selecting a specific item.