naver-blog-research

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

• Third-party content exposure detected (medium risk: 0.30). The skill searches and reads Naver blog posts and search results authored by external blog writers, which are ingested as free text into the agent runtime workflow.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill uses npx -y @nomadamas/k-skill@0 at runtime to pull core instructions and executable helpers from an external npm package repository, constituting a weak/unknown external operational dependency.