korean-slang-writing

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly performs best-effort lookups of public, user-contributed Namu Wiki pages (via scripts/slang_lookup.py and scripts/_slang_http.py) and the SKILL.md requires the agent to read/use those summaries/links when selecting or explaining slang, so untrusted third-party content from namu.wiki can be ingested and materially influence the agent's outputs.