jobkorea-talent-search

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads candidate resumes and profile summaries from JobKorea search results via browser interaction and an AJAX script, which can contain user-submitted free text from job seekers.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill uses npm commands to fetch and execute remote code and instructions from a third-party repository (@nomadamas/k-skill) at runtime, presenting weak/unknown provenance.