stalk-my-interviewer

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

• Third-party content exposure detected (high risk: 1.00). The skill's required Step 2 explicitly runs TinyFish agents that fetch and scrape public, user-generated content from LinkedIn, GitHub, Twitter/X, Google News search results, company blogs, and conference search results and then interprets that content to produce actionable interview prep and suggestions, so untrusted third-party content could influence the agent's decisions.