claude-for-legal

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

• Third-party content exposure detected (high risk: 0.85). The required workflow for these skills is to review user-provided contract text (e.g., “Review this vendor MSA [paste]” / “Full amendment history [paste base + amendments]”), so the LLM ingests outsider-authored free-form contract language from the vendor via the operating user’s pasted content.