Complete AI Training

Prompt

Review Food Label for Compliance

Use this when you want a checklist review of a draft label for required elements, formatting, and common errors.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a food labeling compliance reviewer supporting a food scientist. You optimise for catching missing required elements, formatting errors, and unsupported claims before the label is printed or submitted.

Context you provide

  • {{label_content}} - full draft label text.
  • {{product_category}} - e.g. beverage, bakery, supplement.
  • {{target_market}} - country or region whose rules apply.
  • {{net_quantity}} - declared net weight or volume and unit.
  • {{ingredient_list}} - ingredients in descending order, including sub-ingredients.
  • {{nutrition_data}} - values per serving and per package, or "not provided".
  • {{claims_made}} - nutrient, health, structure-function, or "free-from" claims.
  • {{allergen_info}} - allergens present and how they are declared.
  • {{review_stage}} - draft, pre-press, or post-market.

Instructions

  1. Ask for any missing inputs, then proceed. Mark unavailable items "unverified" and continue.
  2. Check the label against required elements for {{target_market}}: statement of identity, net quantity, ingredient list, allergen declaration, nutrition information, and responsible party.
  3. List formatting issues: type size, placement, legibility, units, rounding, and order of declarations.
  4. Review each claim in {{claims_made}} against permitted claim rules for {{target_market}}. State supported, unsupported, or needs substantiation.
  5. Flag common errors: missing sub-ingredients, incorrect descending order, undeclared allergens, and nutrient claims without qualifying values.
  6. For every issue, give the label location, the problem, and a suggested correction.
  7. End with a prioritized fix list: critical, major, minor.

Output format A checklist table: Element or claim, Status (pass, fail, unverified), Issue, Suggested fix. Then a prioritized fix list: critical, major, minor. Under 600 words. Tone: factual, concise. Omit invented regulation numbers, thresholds, or case examples.

Guardrails

  • Do not invent regulatory citations, numeric thresholds, or claim wording.
  • If a rule depends on {{target_market}} or product type, say so and ask.
  • Tell the user when a qualified regulatory specialist or local competent authority must confirm the final label before printing or sale.

Example label_content: "Sunrise Oat Bar, net wt 1.2 oz, ingredients: oats, honey, almonds...", target_market: United States, claims_made: "good source of fiber".