ai-writing-content

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

• Third-party content exposure detected (high risk: 0.90). This skill's "Step 4: Add research grounding" (Retrieval-augmented content) shows ResearchedWriter using self.retrieve(...) to collect external passages (sources.extend(self.retrieve(query).passages) ) and then passes those sources into WriteSectionWithSources so the agent reads and uses untrusted public web content to ground claims, which can materially influence behavior and enable indirect prompt injection.