enterprise-vpn-attack

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document explicitly instructs on exploiting VPN appliance vulnerabilities to read sensitive files, steal session tokens and credentials, achieve pre-auth remote code execution/web-shells, and pivot to internal takeover — high risk for malicious use.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s workflow description ingests third-party free text only via HTTP responses from outsider-controlled target VPN web endpoints (e.g., login pages, JS/HTML, and SAML metadata fetched with curl), which can contain attacker-controlled content and is read by an LLM for fingerprinting.