Your Postiz pipeline has a creative gap. Here is how to close it.

Postiz publishes. Something still has to create.

If you have automated your publishing queue with Postiz, you have already solved the hard scheduling problem. The friction that remains is upstream: generating a visual that looks intentional, carries your brand fonts and colors, and renders readable text without hallucinating the copy.

Most developers in this position reach for one of three workarounds, and all three have the same failure mode.

Claude reasons. Adly renders. Postiz publishes.

The Adly MCP server exposes Adly's creative engine as a set of callable tools inside any MCP-compatible AI client. Claude can call generate_ad_creative, pass your brand ID, a content brief, and a target aspect ratio, then receive a hosted image URL back. That URL goes straight into the Postiz API call in the same conversation turn.

Adly uses your saved brand profile: your colors, fonts, logo, and any Brand Knowledge photos you have uploaded. The output is a marketing layout with clean rendered text, not a diffusion hallucination. To understand how MCP connects AI assistants to external tools at a protocol level, see what is MCP and how does it work with AI tools.

Add Adly MCP to Claude Desktop in under 2 minutes

Generate your MCP API key from Profile → Adly MCP inside your Adly account. Then open your claude_desktop_config.json and add the Adly server block.

// Minimal config to run Adly MCP alongside any other servers you have.
{
  "mcpServers": {

    // -- Adly creative engine --
    "adly": {
      "command": "npx",
      "args": ["@adly-ai/mcp-server"],
      "env": {
        "ADLY_API_KEY": "YOUR_ADLY_MCP_KEY"
      }
    },

    // -- Add your Postiz API integration here --
    // Postiz exposes a REST API; call it from Claude via fetch or
    // a lightweight MCP wrapper that forwards the Adly image URL.
    "postiz-bridge": {
      "command": "node",
      "args": ["/path/to/your/postiz-mcp-bridge.js"],
      "env": {
        "POSTIZ_API_KEY": "YOUR_POSTIZ_KEY",
        "POSTIZ_BASE_URL": "https://your-postiz-instance.com"
      }
    }

  }
}

Restart Claude Desktop. You will see adly listed under available tools. For a full walkthrough of running Adly from your AI client, see run Adly from Claude ChatGPT and Cursor.

From Claude prompt to Postiz queue: step by step

This is the exact sequence Claude executes when you give it the prompt template in the next section.

This is what Adly produces. Judge it by looking.

The images below were generated by Adly's creative engine via the same job types available through MCP. Clean text rendering, brand-consistent layout, ready to drop into a Postiz queue without touching a design tool.

generate_ad_creative / product_sale

generate_ad_creative / announcement

generate_ad_creative / from_image

Every one of these passed through the same MCP tool call you will use in your pipeline. The text is rendered by Adly's layout engine, not hallucinated by a diffusion model. To see how UGC video assets fit into the same pipeline, see create UGC videos and ad creatives from Claude with Adly.

Adly MCP vs. generic image generation MCPs

This is not about which tool is "better" in the abstract. It is about which tool closes the specific gap in a Postiz pipeline: producing a marketing-ready image with clean text, brand assets, and a layout that does not need a designer to fix it afterward.

Postiz wins on publishing breadth: 30+ channels vs. Adly's Facebook and Instagram publishing. That is the right division of labor. Adly handles the creative step; Postiz handles the distribution. For a detailed look at the Adly MCP tool schema and what each tool does, see Adly MCP for Claude AI marketing creation.

The Claude prompt that runs the full loop

Drop this into Claude Desktop (with Adly MCP and your Postiz bridge connected). Edit the variables in orange. Claude will handle the rest: brand selection, creative generation, polling, approval check, and queue submission.

Adly generating a marketing creative end to end

This is the same creative engine your Claude MCP calls invoke. The output arrives as a hosted URL, ready for Postiz.

What else Adly MCP can generate for your Postiz queue

The generate_ad_creative tool is the starting point. The same MCP connection gives Claude access to the full Adly job library: carousels, UGC-style video ads, Scene Studio multi-scene commercials, product studio videos, and talking character clips. Each returns a hosted URL you can pass downstream.

job_type: announcement

job_type: product_sale

Video outputs from Scene Studio and UGC jobs also return hosted URLs. If Postiz supports video posts on your target channels, the same pipeline applies: Claude calls the Adly video job, polls until complete, and passes the video URL to Postiz. For the full scope of what you can build from Claude, see Adly MCP for Claude AI marketing creation.