atai-newton-fusion-model

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

• Third-party content exposure detected (high risk: 0.75). High: the workflow ingests user-provided file contents (e.g., text/plain text attachments referenced by file_ids in the /query body) as LLM-readable prompt context; this can include outsider-authored free text when the operating user attaches documents they didn’t author.