paper-reading-experiment-design

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow includes “Search or inspect papers” (step 10) and “Read the source artifact and relevant local context” (step 5), which can ingest outsider-authored free text from public web/paper sources into the LLM context via retrieved paper text/abstracts or scraped content.