hypogenic

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests and processes public third‑party content—e.g., cloning public GitHub datasets (https://github.com/ChicagoHAI/HypoGeniC-datasets and Hypothesis-agent-datasets) and preprocessing user-added research PDFs in literature/... via GROBID—and injects those texts into LLM prompt templates for hypothesis generation, meaning untrusted/user-provided content will be read and interpreted by the agent.