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 fetch and read open/public third-party content (e.g., arXiv search URLs and programmatic use of Semantic Scholar / CrossRef / arXiv in references/citation-workflow.md and the Default operating order), and to have the paper-miner agent analyze papers and merge writing signals into a global memory (references/knowledge and paper-miner-writing-memory.md), meaning untrusted external documents are ingested and can materially influence subsequent decisions and tool use.