data-charts-tako

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime calls Tako’s /v1/knowledge_search with source_indexes that can include web, which can fetch and ingest public web page text authored by outsiders into the returned knowledge_cards (e.g., description/sources) that the agent then places into the LLM context.