josh-braun-copywriting

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md instructs the workflow to use ColdIQ MCP/API to fetch/“extract” page content and other enrichment signals at runtime; those signals are not specified as being authored by the operating user, so the LLM context can include outsider-authored free text via mcp__coldiq__fetch_page_content (and similar enrichment calls) from external sources.