article-writing

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

• Third-party content exposure detected (medium risk: 0.30). In skill workflow step 3/4, the agent collects and extracts free text (titles/snippets/headings/questions) from an outsider-authored Google SERP and competitor pages via scripts/collect_serp_outlines.py (DataForSEO organic results plus optional direct/Jina Reader page fetching) and then feeds that untrusted text into the content brief/packets.