humanization-2026

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This content is high-risk: it is explicitly designed to evade anti-bot/behavioral detection (detailed humanization, session persistence, proxy/reputation guidance) and therefore enables large-scale automated abuse (spam, scraping, account takeover) even though it contains no direct backdoor, exfiltration, or credential-theft code.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly directs the agent to read and act on live web page DOM and public social content (e.g., page.evaluate, page.get_by_role / get_by_placeholder, bounding_box, scrolling and the post_to_threads example targeting Threads), so untrusted third‑party page content is ingested and directly shapes mouse/typing/decision behavior.