positive-reply-scoring

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

• Third-party content exposure detected (high risk: 0.90). This skill's fetch script (scripts/fetch-campaign-replies.ts) pulls user-generated email replies from the Smartlead API (https://server.smartlead.ai/api/v1/.../message-history), writes them to JSON, and SKILL.md step 2 explicitly instructs Claude to read and classify those reply bodies—so untrusted third-party content is ingested and directly influences classification and downstream actions.