pace-config

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path ingests OUTSIDER-authored free text from public web sources via adapters like rss (arbitrary urls fetched at runtime) and hackernews/lobsters/youtube (public feeds), and then passes that content into LLM transforms such as llm-filter/llm-summarize/llm-rank/llm-merge when configured.