banking-ai-literature-mapper

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow requires searching and ingesting external banking/adjacent papers, datasets/benchmarks, and regulatory sources (e.g., via public web/API retrieval), which are outsider-authored free-text that the agent will read and include in the LLM context for mapping and evaluation.