REAA Flavor

---

⚠️ This is a modified fork

This repository is an AI-native fork of gitroomhq/postiz-app, built by REAA.

🤖 AI at the core — A governed, multi-provider AI layer powers the whole platform. Bring your own keys across 25 providers — 13 direct model providers (OpenAI, Anthropic, Google Gemini, xAI Grok, Meta Llama, Mistral, DeepSeek, Cohere, Perplexity, Groq, Qwen, MiniMax, Azure OpenAI) plus 12 multi-model hubs & gateways (Amazon Bedrock, Google Vertex AI, OpenRouter, Vercel AI Gateway, Together AI, Fireworks AI, DeepInfra, SiliconFlow, Lightning AI, GMI Cloud, Bitdeer, Vultr) — pick the exact model from an admin screen, and switch providers everywhere without a redeploy. On top of it: on-brand content generation, smart comment replies, brand-voice profiles, a shared prompt library, semantic (RAG) search over your own content, compliance guardrails (prompt-injection / PII / brand-safety / NSFW), and per-org spend caps with a full audit log — every AI entry point scoped, rate-limited, and budget-checked.

Everything else builds around that: persisted multi-channel analytics, a cross-channel comment inbox, campaigns, native polls, 36+ channels, and a security-hardened, self-hosted stack.

Full changelog below (newest first):

[v3.6.0]

• Your own profile page with password and notification controls — Set your avatar, bio, and name in a new settings profile tab, change your password, see your active sessions, and manage email notification preferences — all in /settings/profile.
• Bring your own storage — S3, R2, Backblaze, IDrive — Every organization can mount its own storage provider (S3, R2, Backblaze B2, IDrive e2, or local disk) with 5 GB of free space by default. Migrate files between providers at any time. The old global STORAGE_PROVIDER env var is gone.
  • Storage health & quota dashboard — See which providers are connected (with last-checked timestamp), view your storage usage as a % of quota (warning at 80%+), and drill into usage by folder or provider.
  • Folder-level provider routing — Assign a specific storage provider to a folder, so all uploads to that folder automatically use that provider.
  • Full storage audit log — Every storage operation (mount, unmount, test, migrate, default-folder config) is logged with timestamp, user, and action.
• Connect channels with your own OAuth apps — Configure per-tenant OAuth credentials in the Channels settings tab with a live health dashboard showing connection status — no more global environment variables for every provider's client ID and secret.
• A real media library with folders, tags, bulk actions, and trash — Browse media in a folder tree, drag files between folders, tag and describe files, select multiple assets for batch operations, and search/sort/paginate through your library. Deleted files go to trash (soft-delete); restore them from the trash modal or permanently purge.
• Manage your AI provider and model in-app — Configure AI providers, models, and API keys per organization in the AI settings tab; set up brand voice profiles and a RAG knowledge base (with file upload support) in the Brand tab. OPENAI_API_KEY is deprecated for model resolution — existing keys are auto-migrated on first boot.
• Datatable rebuilds for Teams, Webhooks, Auto Post, Sets, Signatures — Each of these settings screens is now a real datatable with search, sort, pagination, educational empty states, and proper CRUD flows. Campaigns get archive/unarchive toggles and engagement stats.
• Comments inbox fixes, analytics dark mode, calendar improvements — Comments inbox shows proper error and permission states; analytics charts work in dark mode with skeleton loaders; calendar cards reliably display stats and comment badges.
• Get started faster — A new onboarding checklist guides first-time users through connecting channels, configuring AI, creating their first post, and inviting teammates. Channel health monitoring catches token expiry before it breaks publishing.

[v3.5.10]

• It actually starts now — v3.5.9 shipped without booting (a dependency mix-up took the whole API down). The stack boots cleanly again, and a new automated guard makes sure a non-booting build can never ship.
• Posting and your calendar work again — The composer can save, schedule, and publish, your scheduled posts show up on the calendar, and the in-app AI assistant no longer errors on every page.
• Team management you can actually use — Change a teammate's role and click through to their profile, right from settings — not just add and remove.
• Fix failures without leaving the admin screen — Super-admins can retry a failed post or clear a handled error directly from the error log.
• No more "dead app" from rate limits, and a more accessible UI — The hourly request limit is far more generous so normal use never trips it, billing loads, and settings/admin controls are now proper keyboard-accessible buttons.

[v3.5.9]

• Your workspaces stay private — Campaigns, watchlist, and comments are now fully locked to your organization, and every endpoint has proper permission checks — no one can see or touch your data.
• A comment inbox that just works — The inbox loads without 400 errors, comments can be marked read in bulk with org-scoped safety, and a new "Summarize" button lets AI distill the conversation for you.
• Your profile and settings, right where you expect them — A new profile section with your name, bio, and picture lives in settings, your avatar menu in the top bar gives you quick access to settings and logout, and a new Media Providers panel shows at a glance which AI providers power your image, video, and voice generation.
• Admin tools you can actually find — Super-admins see an "Administration" section in the sidebar linking to AI settings, channels, errors, and stats — no more hunting for admin pages.
• More dependable, everywhere — Timezones display correctly on calendars, stats show for every published post, billing forms render completely, and the platform is hardened against memory leaks, race conditions, and N+1 performance pitfalls.

[v3.5.0]

• Analytics that tell you what to do next — See how every channel trends over time, spot your best times to post at a glance, and get a prioritized list of actions to grow faster — and keep an eye on the competitors that matter to you.
• AI that does the busy work — Generate on-brand hashtags, instantly read the mood of your comments, catch compliance problems before you publish, and write new posts in the voice of your best-performing content.
• Engage everywhere from one inbox — Auto-post a first comment for extra reach, run native polls on X and LinkedIn, and reply to comments across all your channels from a single place — now covering even more platforms.
• Plan campaigns, not just posts — Group posts, media, and analytics by campaign, and schedule weeks of content in one move by importing a spreadsheet — with a pre-publish check that flags issues before they go live.
• Faster, safer, more dependable — Your accounts and data are protected by end-to-end encryption and hardened against common web attacks, on a snappier platform that's been re-architected for reliability.

[v3.4.0]

• Use any AI provider you want — You're no longer locked to a single vendor. Choose from 25 providers — 13 direct model providers (OpenAI, Anthropic, Google Gemini, xAI Grok, Meta Llama, Mistral, DeepSeek, Cohere, Perplexity, Groq, Qwen, MiniMax, Azure OpenAI) plus 12 multi-model hubs & gateways (Amazon Bedrock, Google Vertex AI, OpenRouter, Vercel AI Gateway, Together AI, Fireworks AI, DeepInfra, SiliconFlow, Lightning AI, GMI Cloud, Bitdeer, Vultr) — pick the exact model from an admin screen, test the connection, and watch live health badges. Your existing OPENAI_API_KEY keeps working byte-for-byte if you change nothing.
• Stay in control of cost and safety — Set monthly and daily spend caps per org or per use case, get an alert at 80% before you hit the limit, and review every request in a full spend log. Built-in guardrails screen input and output for prompt-injection, sensitive data (PII), brand safety, and NSFW content, with your choice of block, redact, or warn — and a dry-run preview to test a rule before it goes live.
• AI that actually sounds like you — Reusable brand-voice profiles, a prompt-template builder, a shared team prompt library, semantic search across your own content, smart comment replies, and a usage dashboard — so the AI writes in your voice, not a generic one.
• One engine behind everything, bring your own keys — The composer, the agent generator, the chat assistant, and the agent/automation API all run through a single AI provider layer, so switching providers applies everywhere without a redeploy. Teams and self-hosters can plug in their own per-organization keys, and every agent entry point is locked down with scopes, rate limits, and budget checks.

[v3.3.0]

• A calendar that shows performance — Click any post to open a full detail view: key metrics across the top, the entire post thread in one place, a scheduled-vs-published status pill, and real view/like/comment counts right on each card — plus a dedicated settings control to jump straight into editing.
• Conversations on your posts — A new comment-sync foundation keeps the replies on what you publish up to date on a recurring background schedule, tracks what each teammate has already read, and surfaces it all in the app — so you can stay on top of your audience without bouncing between platforms.

[v3.2.0]

• Three more places to reach your audience — Publish to Tumblr, Pixelfed, and PeerTube, taking Postiz to 36 connected channels. Tumblr posts native rich media (images and video) with automatic token refresh; Pixelfed and PeerTube add comment support — all managed from the same dashboard with no extra database setup.
• Correct, not just connected — The new providers shipped with proper formatting (plain-text rendering fixed so posts never leak raw HTML tags) and media-only posting handled correctly, backed by a 64-case provider test suite.

[v3.1.0]

• Real analytics across all your channels — A persisted, multi-channel dashboard that replaces the old one-channel-at-a-time live view. Track genuine period-over-period trends, drill into any channel, metric, or date range with rich charts, and export everything to CSV or JSON — so you can see what's actually working and act on it.
• Built to stay accurate over time — Metrics are snapshotted daily in the background, normalized into a consistent set across providers, and automatically rolled up into long-term weekly history with configurable retention — so collisions between similar platform metrics are eliminated, the numbers stay correct, and the database stays bounded as history grows.

[v3.0.0+]

• Manage every channel from one screen — Set up and control all your social connections from an admin UI instead of editing environment variables, with credentials encrypted at rest. Turn a provider off and your already-connected accounts keep posting, refreshing, and reporting — only new connections are paused. With no database config, everything falls back to environment variables exactly as before.
• Dozens of provider fixes for dependable publishing — A broad hardening pass across all 33 providers fixed real-world publishing bugs (Pinterest and Dribbble token refresh, Bluesky auto-repost thresholds, Telegram bot tokens) and removed unsafe assumptions and import-time side effects — so scheduled posts and token refreshes behave predictably.
• A foundation you can trust — Backed by 1000+ automated tests, a ready-to-run prebuilt image, and a stack migrated to the modern Temporal-based job model with a verified zero-data-loss upgrade path — so self-hosting is stable, predictable, and quick to deploy.

See github.com/reaatech/postiz-app for the full changelog and source.

---

NEW: check out Postiz agent CLI — perfect for OpenClaw and other agents

Your ultimate AI social media scheduling tool

Postiz: An alternative to Buffer.com, Hypefury, Twitter Hunter, etc. Postiz offers everything you need to manage your social media posts, build an audience, capture leads, and grow your business.

Instagram · YouTube · Dribbble · LinkedIn · Reddit · TikTok · Facebook · Pinterest · Threads · X · Slack · Discord · Mastodon · Bluesky

Explore the docs · Watch the YouTube Tutorials

Register · Join Our Discord (devs only) · Public API

NodeJS SDK · N8N custom node · Make.com integration

🔌 See the leading Postiz features

✨ Features

Note: The screenshots below are legacy upstream images and do not reflect this fork's UI. Features such as the persisted analytics dashboard and the channels admin UI are not pictured here.

Intro

• Schedule all your social media posts (many AI features)
• Measure your work with analytics.
• Collaborate with other team members to exchange or buy posts.
• Invite your team members to collaborate, comment, and schedule posts.
• At the moment there is no difference between the hosted version to the self-hosted version
• Perfect for automation (API) with platforms like N8N, Make.com, Zapier, etc.

Tech Stack

• Pnpm workspaces (Monorepo)
• NextJS (React)
• NestJS
• Prisma (Default to PostgreSQL)
• Temporal
• Resend (email notifications)

Quick Start

To have the project up and running, please follow the Quick Start Guide

Postiz Compliance

• Postiz is an open-source, self-hosted social media scheduling tool that supports platforms like X (formerly Twitter), Bluesky, Mastodon, Discord, and others.
• Postiz hosted service uses official, platform-approved OAuth flows.
• Postiz does not automate or scrape content from social media platforms.
• Postiz Users always authenticate directly with the social platform (e.g., X, Discord, etc.), ensuring platform compliance and data privacy.

License

This repository's source code is available under the AGPL-3.0 license.