research-shortform

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

• Third-party content exposure detected (high risk: 0.85). At runtime, the skill’s Platform Scout agent performs WebSearch/WebFetch of public TikTok/Reels pages and captures readable fields like captions/opening text from outsider-authored videos, which are then passed through pattern-extractor/synthesis into the LLM context (indirect prompt-injection risk via scraped free text).