conducting-social-engineering-pretext-call

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content is high-risk dual-use social engineering guidance that explicitly instructs and templates for credential theft (passwords, MFA codes), caller ID spoofing, remote-access escalation, and wire-transfer fraud tactics—behavior that can be directly abused to exfiltrate credentials, gain unauthorized access, or commit financial fraud.

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md Phase 1 "OSINT and Target Research" explicitly instructs using public sources (e.g., theHarvester against LinkedIn, corporate "About Us"/team pages, job postings and other public websites) to gather target data that the agent is expected to interpret to craft pretexts, so untrusted third‑party, user‑generated content can directly influence subsequent actions.