ml-paper-writing

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs the agent to perform literature searches and fetch/verify citations from open/public third‑party sources (e.g., Exa MCP at https://mcp.exa.ai/mcp, Semantic Scholar, CrossRef, arXiv, and general web search/APIs) and requires the agent to read and interpret those results as part of its citation workflow, exposing it to untrusted public content that could carry indirect prompt injection.