cold-call-scripts

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the runtime workflow to call ColdIQ MCP/API functions such as mcp__coldiq__fetch_page_content and mcp__coldiq__extract_post_engagement, whose returned page/post text is outsider-authored free-form web/social content that would be ingested into the agent’s LLM context for scripting.