saml-sso-attack

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit offensive guide describing step-by-step SAML exploitation techniques (XSW, signature stripping, SSO timing enumeration, Golden SAML) that enable credential theft, impersonation, and post‑compromise token forging — clearly malicious in intent.

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

• Third-party content exposure detected (medium risk: 0.30). The skill ingests free text authored by a workflow caller at runtime by accepting attacker-supplied SAMLRequest/SAMLResponse content and decoding/parsing it (e.g., SAMLREQUEST_BASE64 piped into python3 for XML decoding and analysis, and $METADATA_URL metadata fetched and parsed), making indirect prompt injection possible via attacker-controlled XML payloads.