sdr-master-prompts

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

• Third-party content exposure detected (high risk: 0.85). The required workflow references “ColdIQ MCP” calls like mcp__coldiq__fetch_page_content / mcp__coldiq__extract_post_engagement, which would fetch and ingest outsider-authored web/social/free text into the agent’s LLM context at runtime when connected.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent to pull personalization material at runtime via ColdIQ (People Enrichment/Intent/Social endpoints) using the API endpoint https://api.coldiq.com, and that fetched content is used to control the prompts generated.