tweetscape-narrator

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

• Third-party content exposure detected (high risk: 0.90). The skill ingests public, user-generated Bluesky content via the public API and by scraping bsky.app (scripts/ingest.ts and capture-screenshots.ts) and passes those posts verbatim into LLM prompts (scripts/script-narration.ts and scripts/narrate.ts) which directly shape generated narration, scene selection, and downstream recording/rendering, creating a clear vector for indirect prompt injection.