fastgpt-design-skill

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

• Third-party content exposure detected (high risk: 0.90). The skill's documentation (core/dataset/index.md and the listed API routes like projects/app/src/pages/api/core/dataset/collection/create/link and the websiteDataset/externalFileUrl fields) explicitly supports importing/ingesting arbitrary public links, website scraping and API datasets, and those ingested, user-provided third‑party contents are used by the dataset search/RAG flow (packages/service/core/dataset/search/controller.ts and datasetSearchNode) to influence LLM outputs and workflow decisions.