story-bank-builder

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via the skill’s “Reads” path: it ingests user-pasted interview transcripts/case notes/testimonials/win-loss material (explicitly treated as untrusted input), and those pasted texts may contain outsider-authored content that the user did not choose to introduce as their own; additionally, it may fetch public artifacts at runtime with Firecrawl and read the scraped page text.