list-architect

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The content contains deliberate, abuse-oriented guidance: automated scraping of LinkedIn/Google Maps and other sources, instructions to bypass platform limits, webhook pipelines, and building 100K+ cold-email campaigns—patterns that enable large-scale unsolicited data harvesting and outreach with privacy and ToS-evading behavior.

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow can ingest outsider-authored free text at runtime via Clay’s company summary step that prompts the model to “visit this website /company_domain and give me an exhaustive summary,” meaning externally fetched page content (potentially attacker-controlled) is read by the LLM without first selecting a specific trusted item.