startup-validator

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s workflow explicitly instructs using web search and then web_fetch to read full articles from potentially attacker-influenced sources based on the user’s startup idea (outsider-provided free text), meaning the LLM will ingest outsider-authored free text via fetched web pages.