Postiz AI: Setup Guide & AI Features

Postiz AI lets you draft, schedule, and publish social media posts across 30+ channels using AI copilot, image generation, and video clipping. Whether you're running a solo operation or managing a team, Postiz cuts the content creation time from hours to minutes. This guide walks you through setup, first-post workflow, and connecting AI agents like Claude for hands-off scheduling.

What Is Postiz AI?

Postiz AI is a social media scheduler that uses artificial intelligence to draft posts, generate images, and create video clips across 30+ platforms like LinkedIn, Twitter, Facebook, TikTok, and Instagram. You give it a prompt or topic, and the AI writes the content. Then you review, tweak, and schedule it for later. Postiz handles publishing at the time you set.

The tool comes in two forms: Postiz Cloud (SaaS, starting at $29/month) and self-hosted (you run it on your own server). Both support AI features, but the setup and cost model differ. For teams, Postiz removes the "what should I post" bottleneck by letting you batch-create a week's content in an hour.

Postiz has 37,000 GitHub stars and 7 million monthly downloads, meaning a lot of teams already trust it for daily publishing. The open-source nature means you can inspect the code, run it on your infrastructure, or use the managed Cloud version. AI features work the same way on both, though self-hosted gives you more control over which AI models you connect.

Postiz AI Features: Copilot, Images, Video & MCP

Postiz AI includes four main features: AI Copilot for writing posts, AI image generation, AI video clipping, and MCP (Model Context Protocol) for connecting Claude or ChatGPT directly. The feature set varies by plan. Cloud plans include metered monthly allowances. Self-hosted setups let you bring your own API keys and set your own limits.

AI Copilot is the core feature. You describe what you want to post - topic, tone, target audience - and the copilot writes it for you. You can refine it, ask for variations, or ask it to generate hashtags and thread ideas. It works on all plans.

Image generation creates social media graphics to match your posts. On Cloud, standard plans may include zero AI-generated images per month, while Team plan gives you 100 per month. Self-hosted users connect their own OpenAI API key (OPENAI_API_KEY) and pay per image directly to OpenAI.

Video clipping turns long-form content into short TikTok-style clips. Team plan allows 10 video clips per month. Higher tiers offer more capacity. This feature requires external API keys on self-hosted.

MCP (Model Context Protocol) is the newest feature. It lets you connect Postiz to Claude or ChatGPT and ask the AI agent to draft and schedule posts directly. Instead of logging into Postiz and clicking buttons, you say: "Write 5 LinkedIn posts about DevOps best practices and schedule them for Monday through Friday at 9am." Claude does the work, then Postiz publishes them.

Self-Hosted vs Cloud: Which AI Features Do You Get?

Cloud plans include built-in AI with metered monthly allowances. Self-hosted setups require you to bring your own API keys (like OPENAI_API_KEY for image generation). Neither approach charges you less per feature; Cloud bundles costs, self-hosted lets you pick which models and pay per use. The trade-off is convenience vs control.

On Postiz Cloud, you sign up, choose a plan starting at $29/month, and AI features are ready to use. No API keys to find or configure. You get monthly allowances that reset on the same date each month. Team plan includes 100 AI images and 10 video clips. See postiz.com/pricing for exact tier pricing and features.

Self-hosted Postiz gives you more flexibility. You install Postiz on your own server (using Docker, a VPS, or managed Postiz hosting through Opsily). Then you bring your own API keys: OpenAI API key for image generation and post writing, or Anthropic key for Claude. You pay per-request, not per-month. A high-volume creator might spend $50-300/month on API calls depending on usage.

The MCP feature (Claude/ChatGPT integration) works on both. On Cloud, the copilot itself uses API credits. With self-hosted, you connect Postiz to Claude or ChatGPT, and the AI uses YOUR account credits. This is actually cheaper for power users because you're not paying Postiz's markup on token usage.

Step 1: Setting Up Postiz AI on Cloud (5 Minutes)

Go to postiz.com, create a free account, choose the plan, and connect one social channel. Postiz will ask for OAuth permissions from that channel (LinkedIn, Twitter, etc.). Once connected, you can start creating posts immediately. No API keys needed.

Here's the exact flow. First, visit postiz.com and click "Start Free." Enter your email and password. Postiz will show you available plan options. Choose one. Standard or lowest-tier plan is a good starting point to test the copilot.

Next, Postiz asks you to connect your first social channel. Click "Add Channel" or "Connect LinkedIn" (or Twitter, Facebook, TikTok - you pick). Postiz redirects you to LinkedIn to authenticate. You'll see a prompt: "Postiz is asking for permission to post on your behalf." Click "Authorize." LinkedIn redirects you back to Postiz. Done. You're now connected and can publish to that channel.

If you're on a paid plan with image and video generation, those features are already enabled in the composer. You'll see an "Ask AI" button next to the text editor. Lower-tier users may see it but with features grayed out. No configuration needed. The AI knows your API key because Postiz has already set it up for you.

Step 2: Drafting Your First Post with AI Copilot

Click the "Compose" button in Postiz. Write your prompt in the text field: describe the topic, target audience, and tone. Click "Ask AI" or "Generate with AI." The copilot writes the post. Review it, edit if needed, then add an image, set a schedule, and publish.

Open Postiz and click "Compose Post" or the "+" button. A modal opens with a text editor. This is where you write your post or prompt the AI.

To use the copilot, type a prompt like: "Write a LinkedIn post about time management for founders. Keep it professional but approachable. 150 words. Include a hook about productivity paradoxes."

Click "Ask AI" (or the sparkle icon). Postiz sends your prompt to its AI model. In 5-10 seconds, you'll see a draft in the composer. It reads like a real post, not a bot-generated mess.

Now you edit. Click inside the generated text and make changes. Delete a sentence. Rephrase a joke. Add an emoji. Postiz treats it like a normal text field. When you're happy, click "Add Image" if you want a graphic. Click "Generate AI Image" and describe what you want: "Professional photo of a founder at a desk, laptop, morning light, laptop screen showing calendar." The copilot generates an image in 30 seconds.

Next, set when to publish. Click "Schedule" and pick a date and time. For now, pick tomorrow at 9am. Click "Schedule Post." Postiz confirms the post is queued. Check your calendar (in Postiz) and you'll see the post scheduled for tomorrow at 9am on LinkedIn. That's it. One post, 10 minutes.

Step 3: Connecting Claude to Postiz (Advanced)

MCP (Model Context Protocol) is a new standard that lets Claude or ChatGPT control Postiz directly. You install Postiz as an MCP server in Claude Desktop or connect it via ChatGPT plugin. Then you can ask Claude: "Create 5 LinkedIn posts about DevOps best practices and schedule them for this week at 9am." Claude writes and schedules them all.

MCP sounds complex but it's not. Here's why it matters: instead of you going into Postiz, writing prompts, waiting for drafts, editing, and scheduling - Claude does all that while you read the output. You stay in Claude and ask it to do the work. It's like giving Claude a "Postiz tool" to use.

To set it up, you need:

• Claude Desktop app (free from anthropic.com)
• Postiz self-hosted running on your machine or server (or a Postiz Cloud account; MCP works with both)
• The Postiz MCP server configuration

The setup is technical, but Postiz docs (docs.postiz.com) walk you through it. Roughly: you download the MCP server, add it to Claude's config file, restart Claude, and you're done. Within Claude, you'll see a "Postiz" tool.

Once it's set up, you can ask Claude in plain English: "Write 5 LinkedIn posts for a DevOps SaaS company. Each post should cover one topic: Docker best practices, Kubernetes cost optimization, CI/CD pitfalls, monitoring strategies, security hardening. Make each 200 words. Professional tone. Include relevant hashtags. Schedule them for next Monday through Friday at 9am." Claude drafts all five, you review them in the message, and Claude schedules them to Postiz. No manual clicks.

Practical Weekly Workflow: From One Prompt to Seven Scheduled Posts

Here's a real example: a SaaS founder spends 60 minutes on Monday morning to create a week of LinkedIn content. She writes one prompt to the AI: topics, audience, tone. The AI generates seven drafts. She spends 20 minutes tweaking them. Then schedules all seven for 9am Monday through Sunday. Done.

Let's say you're a DevOps SaaS founder and you want to build presence on LinkedIn. Your audience is engineering leaders. Your topics: container orchestration, cost reduction, monitoring, and team workflows. Your goal: one post per day for a week.

Monday 8am, you open Postiz (or Claude if using MCP). If using Postiz directly, you click "Compose." You write:

"Generate 7 LinkedIn posts for engineering leaders at mid-size SaaS companies. Topics: Kubernetes cost optimization, Docker security best practices, monitoring observability, CI/CD efficiency, team on-call workflows, infrastructure automation, cloud FinOps. Each post should be 150-200 words. Professional but not stiff. Start with a relatable hook or question. End with a call to action (e.g., 'What's your biggest challenge here?'). Include relevant hashtags."

Click "Ask AI." Postiz generates the seven posts as a batch. You copy them one by one into the composer, tweak as needed, add images (AI-generated), and schedule them for 9am each day of the week.

If using MCP, you ask Claude: "Create 7 LinkedIn posts for engineering leaders on [same topics]. Professional, 150-200 words each, relatable hooks, calls to action, hashtags. Schedule them for Monday-Sunday at 9am on Postiz." Claude generates all seven, shows them to you in a nice format, and handles scheduling. You approve or ask for changes, then Claude confirms they're in Postiz. Total time: 30 minutes, not two hours.

The math: 7 posts times 3 hours each (research, writing, editing, scheduling manually) equals 21 hours. With AI: 30 minutes to 1 hour. You save 20 hours a week. That's two full work days. If you're paying yourself $50/hour, that's $1,000/week in time savings alone.

Gotchas, Limits & When to Self-Host vs Cloud

Image and video generation quotas reset monthly on the same date. Once you hit the limit, you can't generate more until next month. Self-hosted Postiz requires managing API keys and server maintenance. Cloud is simpler but costs money. Self-hosted costs zero for the software but you pay for API usage and hosting.

First, the quotas. On Team plan, you get 100 AI images per month. If you use them all on day 5, you're done until day 1 of next month (when the counter resets). The same goes for video: 10 per month. This is by design - Postiz wants to prevent someone from running unlimited generations and racking up massive API costs. For higher volume, upgrade to a plan with more allowance or go self-hosted with your own API keys.

Second, self-hosted setup requires some technical chops. You need to install Docker or provision a VPS, run Postiz in a container, configure reverse proxy (so Postiz has a web address), set up SSL (HTTPS), and manage backups. If you're not comfortable with command line, pay someone $500-2,000 to set it up, or use managed Postiz hosting (which handles all that for you). Cloud avoids this headache entirely.

When to choose each option:

Choose Cloud if: You want to start today, no setup hassle. You post fewer than 30 times per month (so quotas don't hurt). You're not at massive scale. You want daily backups and uptime guarantees.

Choose self-hosted if: You're posting 50+ times per month and Cloud quotas would block you. You want full control over your data. You're willing to handle or pay for maintenance. You have technical staff or a budget for DevOps help.

One more gotcha: self-hosted requires the right infrastructure. Postiz runs on Linux (Ubuntu recommended). Windows servers won't work out of the box. You need at least 2GB RAM, 10GB disk space, and a domain name pointed at your server. If you're renting a VPS, this costs $10-30/month. Add it to your total self-hosted cost.

Frequently Asked Questions

Can AI agents like Claude or ChatGPT post to social media through Postiz?

Yes, through MCP (Model Context Protocol). You install Postiz as an MCP server in Claude Desktop or connect it to ChatGPT, and then Claude can read your Postiz account, draft posts, and schedule them. It requires setup but works smoothly once configured. ChatGPT support is newer and may require a plugin.

What social media channels does Postiz support?

Postiz supports 30+ channels: LinkedIn, Twitter/X, Facebook, Instagram, TikTok, Threads, Bluesky, Pinterest, Reddit, YouTube, Mastodon, and many others. You can connect multiple channels to one Postiz account and schedule to all of them at once, or tailor different posts to each platform.

Is Postiz free?

Postiz Cloud has entry-level options to test it, with paid plans starting at $29/month. Self-hosted Postiz is free software (open-source), but you pay for hosting infrastructure and API keys separately.

How do I set up Postiz AI image generation?

On Cloud, image generation is available on paid tiers: Team plan includes 100 images per month. You click "Generate AI Image" in the composer and describe what you want. On self-hosted, you set your OPENAI_API_KEY environment variable and images are generated via OpenAI (you pay per image).

What is Postiz compared to Buffer?

Postiz vs Buffer shows key differences in pricing and features. Postiz Team plan includes 100 AI images and 10 video clips per month. Postiz also supports 30+ channels with no per-user fees. Buffer has stricter per-member licensing and fewer channels. Postiz is also open-source, so you can self-host; Buffer is SaaS-only.

Do I need an OpenAI API key to use Postiz AI?

On Cloud, no. Image and video generation are included in the plan; Postiz manages the API key for you. On self-hosted, yes: you provide your OPENAI_API_KEY to enable image generation. You don't need one for the copilot (text writing); the copilot uses Postiz's own model.

How much does Postiz AI cost?

Cloud plans start at $29/month. Team plan includes 100 AI images and 10 video clips monthly. Self-hosted: free software, but you pay for hosting ($0-50/month for a basic VPS) and API usage ($0-300/month depending on volume). No per-user fees on any plan.

What are the AI image generation limits by Postiz plan?

Team plan: 100 AI images per month. Video clipping limits: Team has 10 per month. Higher tiers offer more or unlimited capacity. Limits reset on the date you signed up each month.

The Bottom Line

Postiz AI removes the hardest part of social media: writing enough content consistently. Whether you use Postiz Cloud (simple, metered) or self-hosted (flexible, hands-on), the AI copilot, image generation, and MCP integration save you hours per week. Start with the lowest tier to test the copilot. If you're posting more than a few times per week, upgrade or self-host. The MCP feature is for advanced users who want Claude or ChatGPT to handle scheduling directly.

Ready to automate? Start a free Postiz account today or explore managed Postiz hosting to skip the setup entirely.