browsing-bluesky

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

• Third-party content exposure detected (high risk: 0.90). This skill directly fetches user-generated, public Bluesky content from external endpoints (e.g., https://api.bsky.app and the Jetstream firehose wss://jetstream1.us-east.bsky.network) via functions like search_posts, get_user_posts, get_feed_posts, sample_firehose and analyze_account, and the agent is expected to read and act on that content (trending selection, account categorization), so untrusted third-party content can materially influence its decisions.