marketing-pipeline-share-ai-content

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

• Third-party content exposure detected (high risk: 0.95). The required runtime workflow calls crawlRecentNews() which fetches public web/news content via RapidAPI (https://api.rapidapi.com/news/search), then concatenates n.title and n.content into researchData and injects it into the LLM prompt in generateContent()/generateContentOpenAI() as free text.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill calls the RapidAPI news endpoint at runtime (https://api.rapidapi.com/news/search) via axios to fetch articles that are injected into the AI prompt as "researchData", so external content directly influences model instructions and is a required runtime dependency.