ask_nutrition
ChatGPTAnswer a nutrition question with citations from AllNutrition's vetted library. Industry-funded and biased studies are filtered out before they reach the answer. Use this for any nutrition, diet, or supplement question where you need a trustworthy, citable answer rather than a guess from the open internet. Returns exactly these fields, and no others: question echo of the question asked answer markdown answer grounded in the cited sources evidence_strength strong | moderate | limited | insufficient — quality of the retrieved sources (count of primary studies or syntheses of any design and their trust), not a body-of-evidence verdict consensus_level high | moderate | mixed | low last_updated most recent publication date among the cited sources deep_research whether the deep-research pipeline produced this answer sources[] id, title, url, publisher, publisher_short_name, evidence_level, trust_score, published_at, passage, source_basis, trust_basis `
passage is a short AI-generated summary of the supporting section — verbatim source text is never returned. source_basis is full_text or abstract (the answer read only the paper's abstract, so method details may be missing); trust_basis says which of the two trust_score was assessed from. No field contains personal data or anything about the caller's AllNutrition account. The question and answer are saved to the caller's own private Ask history in the AllNutrition web app by default, so they can revisit them later. That record is visible only to their account and never shared; saving can be disabled per call (save_session=false) or account-wide in the web app under Settings → API keys. Args: question: The nutrition question, in plain English (max 1000 characters). published_after: Optional ISO-8601 cutoff (e.g. 2020-01-01T00:00:00Z`) to bias the answer toward more recent evidence. deep_research: Expand the question into multiple targeted sub-queries and synthesize a comprehensive, sectioned research report instead of a standard answer. SLOW BY DESIGN: typically 30-120 s — that is normal, not a hang or an outage, so keep the call open and wait for it. Enable it when the user asks for deep research, a comprehensive report, or an in-depth review. If it errors, retry with deep_research=true — do not downgrade to a standard ask, which answers a shallower question than the one that was asked. save_session: Set false to skip saving this question to the caller's private Ask history. Omit to use the default (save, unless the account has turned saving off in its settings).
