ce-spec

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

• Third-party content exposure detected (high risk: 0.80). The skill's Phase 2 "Run external research conditionally" (and its listed subagent roles like best-practices-researcher and framework-docs-researcher) explicitly fetches and cites external URLs and directs the agent to read and use those public third‑party sources to shape the spec, so untrusted web content could indirectly inject instructions that change behavior.