algo-social-sentiment

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests and analyzes public social-media user-generated content (e.g., "從 Twitter/X 和 Dcard 爬取了 450 則留言" in examples/sample_scenario.md and SKILL.md's "Analyzing sentiment in social media posts, tweets, or reviews"), which the agent reads and uses to drive classifications, escalations, and recommended actions, exposing it to potential indirect prompt injection.