categorizing-bsky-accounts

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

• Third-party content exposure detected (high risk: 1.00). The skill fetches public, user-generated Bluesky content via the script's API calls (e.g., get_author_feed and get_all_following in scripts/bluesky_analyzer.py which call https://public.api.bsky.app/xrpc) and explicitly sends bio+posts/keywords to Claude for categorization (described in SKILL.md), so untrusted third-party text can directly influence the agent's decisions and actions.