Rule, applied consistently everywhere via a new isFeatureAvailableAnyTier() helper: if a feature is enabled on at least one tier, it stays visible for locked-out viewers with a small "Pro" badge and opens an upgrade prompt on click; if a feature is disabled on every tier, it hides entirely, since there's no upgrade path to point at.
Covers: recipe variations, meal/drink pairings, nutrition estimation, Markdown export (5 call sites), weekly nutrition, import from URL, import from photo, and the Instacart grocery-delivery menu item. Previously inconsistent — some hid outright, one showed a lock icon overlapping its own icon.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Extends the existing feature-flags system (previously 3 keys, all
default-enabled) with 6 more: recipe_import_url, recipe_import_photo,
nutrition_estimation, markdown_export, weekly_nutrition,
grocery_delivery. Each FEATURE_DEFINITIONS entry now carries its own
defaultEnabled -- the new 6 default to false, the original 3 stay
true -- so no migration/seed was needed for the "off by default"
requirement, just a per-key fallback instead of a blanket true.
Gated server-side (requireFeatureEnabledResponse, a new shared
helper avoiding six copies of the same try/catch) on: import-url,
import-photo, nutrition POST estimate, bulk markdown export, weekly
meal-plan nutrition GET, Instacart export. Gated client-side by
hiding the trigger entirely (not just disabling) on every page that
renders one: recipe detail (meal/drink pairing buttons, nutrition
panel's estimate button, markdown export), recipes list (import-URL
button, including the OS Share Target auto-import path), new-recipe
page (photo import), meal-plan page (markdown export, weekly
nutrition bar), shopping-list/collection/pantry pages (markdown
export), shopping-list page (Instacart button, now gated by both the
existing env-var check AND the tier flag).
Also: BYOK section on Settings -> AI now hidden entirely for
non-BYOK users (previously showed a locked-and-teased notice, same
inconsistency the Model Prefs fix closed yesterday). Language
switcher shows a flag icon (FlagGB/FlagFR, moved from
components/marketing to components/shared so both the logged-in
settings switcher and the logged-out marketing one can use it)
instead of plain "English"/"Français" text-only options.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Admins can now disable specific AI features per tier from Admin > Tier
Limits — new feature_flags table (feature x tier -> enabled, defaulting
to true so adding a new gated feature never needs a backfill).
Covers recipe variations, drink pairing, and meal pairing to start.
When disabled for a user's tier, the button stays visible (with a small
lock badge) but opens an upgrade dialog instead of running; the API
route rejects the call server-side either way (requireFeatureEnabled,
re-reads tier from the DB rather than trusting the session's cache,
same rationale as checkAndIncrementTierLimit).
The upgrade dialog is informational only — no Stripe checkout exists
yet (STRIPE_PLAN.md is still just a plan) — its CTA links to /support
prefilled as an upgrade-interest suggestion.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Arnaud