deep-research

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

• Third-party content exposure detected (high risk: 0.85). The skill is a “deep research” tool that (per its purpose/workflow) performs runtime web searching/reading and then synthesizes that content into an LLM report, so outsider-authored free text from public web pages is ingested into the model context via the research/search→read→synthesize path.