job-posting-match

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

• Third-party content exposure detected (medium risk: 0.30). scripts/job_posting_match.py fetches and parses public search result HTML from JobKorea and Saramin (e.g., search_jobkorea/search_saramin -> fetch() -> parse_jobkorea/parse_saramin), ingesting outsider-authored free text from those pages without selecting a specific item beyond the search results.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires running npx to fetch and execute the remote npm package @nomadamas/k-skill at runtime (e.g., "npx -y @nomadamas/k-skill@0 instruct job-posting-match"), which downloads and runs external code that controls the skill's instructions.