launch-positioning-builder

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly instructs runtime web research and fetching competitor pages (public web content) and optionally scraping G2/Capterra reviews, which would ingest outsider-authored free text into the agent’s LLM context via web_search/fetch_webpage/review-scraper.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches competitor webpages at runtime (see "Search: "[competitor name]" site:[competitor-url]" and "Fetch: competitor homepage, pricing page, about page"), meaning external page content (site:[competitor-url]) will be injected into the agent's context and directly influence prompts/outputs.