ct-alpha

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests public X/Twitter content via the X API (see SKILL.md prerequisites and lib/api.ts searchRecent/getThread), parses and extracts user-generated text/URLs (lib/extract.ts), and uses those signals to rank results, detect raids, and drive follow-up tool suggestions (lib/tweetrank.ts and resources/tool-discovery.md), so untrusted third‑party content can materially influence agent actions.