fda-ftc-rules

name	fda-ftc-rules
description	Health claim rules, price truth, promotions compliance
plugin	git-plugin-amc

Skill · FDA / FTC Rules

F&B Content Engine · compliance/fda-ftc-rules.md

─────────────────────────────────────────────────

Purpose

Every post must pass this compliance gate before publishing. This file defines the RED (hard block) and YELLOW (owner sign-off required) rules based on FDA, FTC, and general advertising standards.

Update cadence: Review monthly + immediately after any platform policy update or regulatory change.

---

🔴 RED Rules — Hard Block (fix or kill the post)

R1 · Health & Medical Claims (FDA + FTC)

Forbidden phrases — any of these trigger a RED:

• "boosts immunity" / "immune-boosting"
• "good for digestion" / "aids digestion"
• "lowers cholesterol" / "heart-healthy" (unless medically certified)
• "anti-inflammatory"
• "detox" / "cleansing"
• "cures" / "prevents" / "treats" (any health condition)
• "superfood" (unless the claim is explicitly hedged)
• Any claim that a dish improves, maintains, or affects a health condition

Replace with sensory / experiential language:

Forbidden	Allowed
"boosts immunity"	"warming and comforting"
"good for digestion"	"light and satisfying"
"anti-inflammatory"	"made with fresh ginger and spices"
"detox"	"clean, simple ingredients"
"healthy" (as medical claim)	"fresh", "made daily", "no MSG" (if true)

---

R2 · Allergen Non-Disclosure

If a dish contains any of the 9 major allergens, the post MUST include disclosure.

The 9 major allergens (FDA 2023): milk · egg · fish · shellfish · tree nuts · peanuts · wheat · soy · sesame

Rule: If a promoted dish contains any of these, the post must either:

• State "contains [allergen]" in the caption/description, OR
• Link to the menu allergen page

Check [[compliance/allergen-gate]] for the brand's dish-specific allergen table.

---

R3 · Price & Promotion Truth (FTC)

• Stated price MUST match the menu / order page at posting time
• "Limited-time" deals must have an explicit end date (e.g., "until Sunday" or "this week only")
• "Up to X% off" must list a representative sample at that discount level
• Never imply a deal is exclusive when it's available everywhere
• Any post claiming price > $50 or deal > 30% off → automatic Tier 1 escalation
• First post of a new dish → automatic Tier 1 escalation (verify price before publish)

---

R4 · Alcohol Content

• If a dish contains alcohol (residual ABV > 0.5% from cooking counts), do NOT target:
  • Under 21 audience (US)
  • Under 18 audience (CA)
• TikTok: No alcohol promotion at all on most accounts — check account settings
• Instagram: Age-gated audiences only for alcohol content
• Google Business: Limited alcohol promotion allowed
• Soju, beer, cooking wine cocktails → always flag for review

---

R5 · False Reviews & Solicitation

• Never offer discounts, free items, or incentives in exchange for reviews (Violates Google policy + FTC endorsement guidelines)
• Never write, paraphrase, or repurpose customer reviews as original brand content without: (a) written consent from the customer + (b) clear attribution
• Never post fake reviews or instruct others to

---

R6 · Image Rights

All photos used in posts must be one of:

• (a) Shot by the brand (owned)
• (b) Licensed (keep license documentation in Media Index Bitable (attachments))
• (c) Re-posted with explicit written permission from the original poster

Customer photos: Use ONLY with DM consent. Save consent screenshot in Media Index Bitable (attachments). Stock photos: Only with commercial license. Note source in Media Index Bitable. AI-generated images: Only if brand owner has explicitly enabled mcp.image_gen in SOUL.md config.

---

🟡 YELLOW Rules — Owner Sign-Off Required

Any of the following requires owner approval before publishing. Trigger Tier 2 escalation via [[operations/owner-approval]].

Trigger	Why it's sensitive
Comparison to a named competitor	Defamation risk, brand positioning
"First / only / best in [neighborhood]"	Must be provable
Ingredient origin claim ("imported from Sichuan")	Must be documentable
Heritage claim ("authentic", "traditional", "ancient recipe")	Must be defensible
Charitable or political associations	Brand alignment risk
Holiday content tied to specific religion or community	Cultural sensitivity
Any content about a food safety incident (even resolved)	Legal sensitivity

---

🟢 GREEN — Always Safe to Publish

No flag needed for any of the following:

• Sensory descriptors: sizzling, steaming, fresh, hand-pulled, wok-fired, slow-braised
• Real prices and portion descriptions (when verified against current menu)
• Customer experience stories (with consent)
• Behind-the-scenes kitchen content
• General family / community / neighborhood themes
• Deal alerts with explicit end dates and verified prices
• Dish introductions with cultural context (no health claims)

---

Compliance Gate Workflow

1. Repurpose Agent finishes drafts
2. Scan for RED keywords (see R1 list above)
3. Check structured fields: price verified? allergen disclosed? deal has end date?
4. RED hit → block post + create Tier 1 entry in [[operations/owner-approval]]
5. YELLOW hit → flag in ownerreview Lark Doc, hold for owner sign-off (Tier 2)
6. GREEN → pass to [[localization/bilingual-gate]]