interview-coach

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

• Third-party content exposure detected (medium risk: 0.30). In the workflow, the agent ingests outsider-authored free text when “Job posting: URL or pasted text” is provided by the user (and is then used to generate prep content), plus it may ingest untrusted web page content via optional browsing in step 2 (“inspect the website, careers page, recent news, reviews”).