soroush.asadi
3fc7bf2b97
Build backend images / build content-svc (push) Failing after 3m39s
Build backend images / build file-svc (push) Failing after 52s
Build backend images / build gateway (push) Failing after 58s
Build backend images / build identity-svc (push) Failing after 1m21s
Build backend images / build notification-svc (push) Failing after 1m0s
Build backend images / build render-svc (push) Failing after 58s
Build backend images / build studio-svc (push) Failing after 55s
AI SEO content generator - content-svc: per-tenant OpenAI config (ai_settings) + /v1/ai endpoints (settings GET/PUT, seo-post) with SEO-expert prompt → structured article - admin UI to configure token/base-url/model and generate + save as blog - configurable base URL for restricted networks Full data-driven admin panel - generic /api/admin/resource proxy + reusable AdminResource component - categories/tags/fonts/blogs (CRUD), users (list + ban), plans/slides - AI content section; nav + i18n i18n localization sweep - localized 116 user-facing + studio/editor components to next-intl (fa+en) under the auto.* namespace; merge tooling in scripts/merge-i18n.js Branding + assets - Monoline F logo (LogoMark + favicon) - offline SVG placeholder generator (/api/placeholder), dropped picsum.photos Fixes - JWT issuer mismatch on content/studio (flatrender → flatrender-identity) - missing role claim → [Authorize(Roles="Admin")] now works (RBAC) - Secure cookies broke HTTP sessions → gated behind AUTH_COOKIE_SECURE - Radix RTL via DirectionProvider (right-aligned menus in fa) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
21 lines
1013 B
SQL
21 lines
1013 B
SQL
-- =====================================================================
|
|
-- CONTENT SCHEMA — Part 17: AI settings (per-tenant OpenAI config)
|
|
-- Stores the OpenAI (or OpenAI-compatible) API credentials used by the
|
|
-- AI SEO content generator. One row per tenant; api_key is stored as-is
|
|
-- (self-hosted) and never returned in full to clients (masked in the API).
|
|
-- base_url is configurable so deployments behind a proxy / in restricted
|
|
-- networks can point at a reachable OpenAI-compatible endpoint.
|
|
-- =====================================================================
|
|
|
|
SET search_path TO content, public;
|
|
|
|
CREATE TABLE IF NOT EXISTS ai_settings (
|
|
tenant_id UUID PRIMARY KEY,
|
|
provider TEXT NOT NULL DEFAULT 'openai',
|
|
api_key TEXT,
|
|
base_url TEXT NOT NULL DEFAULT 'https://api.openai.com/v1',
|
|
model TEXT NOT NULL DEFAULT 'gpt-4o-mini',
|
|
enabled BOOLEAN NOT NULL DEFAULT FALSE,
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
|
|
);
|