ml-paper-writing

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to search and ingest content from public third-party sources—e.g., Exa MCP (https://mcp.exa.ai/mcp), Semantic Scholar, CrossRef, arXiv, and arbitrary web search/APIs—and requires the agent to read/verify those external papers/BibTeX as part of its citation workflow, exposing it to untrusted third-party content that could carry indirect prompt injection.