How to Build a Multi-Agent AI WordPress Publishing Team

I typed a sentence into a chat box last week. “Create an article about AI trends, find some images and save it as a draft.”

Then I saw three AI agents working together in real time. One searched Pixabay for relevant images. Another wrote a full article with proper titles, embedded images, and image credits. A third party pushed the whole thing as a draft to WordPress, ready for my review.

The whole process took about 90 seconds. No code. No API hassle. Just a single command in natural language.

In this tutorial, I’ll walk you through building exactly this system using Aptevaan open-source AI agent framework. Ultimately, you have a working team of three specialized agents (a coordinator, an image agent, and a WordPress agent) who can automatically research images, write articles, and publish drafts to your WordPress site. In a landscape where staying visible means adapting to AI-driven workflows, it’s important to have this kind of infrastructure in your toolkit.

If you manage content for clients, run a publication, or just want to see what multi-agent AI workflows look like in practice, this one is for you. It’s one of the most practical uses of AI tools for WordPress content creation that I’ve come across.

What you need

Before we start building, make sure you have these ready:

• A Apteva account (free account or local installation) – You can run Apteva locally or use a hosted account to get started faster
• A WordPress site with administrator access
• A Pixabay API key (also free)
• Basic understanding of the WordPress admin dashboard

Estimated time to complete: ~20 minutes.

Step 1: Install the WordPress MCP plugin

MCP (Model Context Protocol) is how your AI agents talk to external services. Think of MCP servers as translators that sit between your agents and the tools they need. As AI adoption accelerates across the WordPress ecosystem, this type of integration becomes essential. For WordPress we use a plugin called Stifli Flex MCP making your site an MCP-compliant endpoint.

Log in to your WordPress admin dashboard and navigate to Plugins > Add New. Search for ‘Stifli Flex MCP’ and then click Install now And Activate.

That’s it for the installation. One plugin and your WordPress site is ready to receive instructions from AI agents.

Step 2: Configure MCP authentication

Once the plugin is activated, you will need to set up secure authentication for Apteva to communicate with your site.

Go to Settings > Stifli Flex MCP in your WordPress admin. You’ll see an authentication section that explains how application passwords work. Click Create application password and keep it in a safe place.

Please note two things here. The JSON-RPC 2.0 endpoint is your base URL. The Endpoint for Claude/ChatGPT contains your username and application password baked into the URL, making configuration in Apteva easy.

The formatted URL will look something like this:

https://yourname:YOUR_APP_PASSWORD@yoursite.com/wp-json/stifli-flex-mcp/v1/messages

Keep this URL handy. You’ll need it for the next step.

Step 3: Connect your MCP servers in Apteva

Now we are switching to Apteva. You’ll need to register two MCP servers: one for WordPress (so agents can create posts) and one for Pixabay (so agents can search for images).

Add the WordPress MCP server

In Apteva, navigate to MCP in the left sidebar and click + Add server. Select HTTP endpoint as the server type.

Fill in the details:

• Name: WordPress
• Domain: Global (all projects)
• URL: Paste the endpoint URL from step 2
• Username: Your WordPress username
• Password: Your application password

Click Add server and you are connected.

Add the Pixabay MCP server

Click + Add server again, but this time select npm package as the server type.

Configure it:

• Name: Pixabay
• Domain: Global (all projects)
• npm package: pixabay-mcp
• Environmental variables: Add PIXABAY_API_KEY with your API key as the value

Verify your servers

Navigate back to the MCP page to confirm that both servers are running. You should see green status indicators next to each status indicator.

Tip: If a server shows as offline, double-check your login information. For Pixabay, check if your API key is valid. For WordPress, make sure the Stifli Flex MCP plugin is still activated and your application password has not been revoked.

Step 4: Create the coordinator agent

With your MCP servers up and running, it’s time to build the agent team. We’ll start with the coordinator, the agent who receives your requests and delegates work to the others.

In Apteva, navigate to Agents in the sidebar. If this is a new project, you will see an empty status.

Click + New agent and configure it:

• Name: Coordinator
• Project: WordPress team
• Provider: Anthropic
• Model: Claude Haiku 4.5 (fast)
• System prompt: You are a coordinator agent.

Under Features, enable Memory (for persistent reminder during conversations) and Multi-agent (so it can communicate with fellow agents). Set the Multi-Agent mode to Coordinatormeaning it orchestrates and delegates rather than receives tasks.

Save the agent. You will see it appear in your project.

Step 5: Create the Work Agents

Now you need two work agents, each specialized for a different part of the publishing workflow.

The image agent

Click + New agent and configure:

• Name: Image agent
• Project: WordPress team
• Provider: Anthropic
• Model: Claude Haiku 4.5 (fast)
• System prompt: You are an image researcher agent.

Enable these features: Memory, Vision (to process and analyze images), MCP (to access external tools), and Multi-agent. Set the Multi-Agent mode to Employee (receives delegated tasks).

When you scroll down to the MCP Servers section, select the Pixabay server for this agent.

The WordPress Agent

Create another agent:

• Name: WordPress agent
• Project: WordPress team
• Provider: Anthropic
• Model: Claude Haiku 4.5 (fast)
• System prompt: You are a wordpress agent.

Enable Memory, MCPAnd Multi-agent. Set Multi-agent mode to Employee. Point the WordPress MCP server.

This agent does not need Vision because it works with text content. Keep the possibilities focused.

Step 6: Check your agent team

Once all three agents have been created, start them all and check if the team is running.

Here’s a quick summary of how the possibilities break down:

The coordinator does not have direct access to the tools. It only knows how to delegate. The employees have limited, specific capabilities. This separation is purposeful. It keeps every agent focused and reduces the chance of mistakes.

Step 7: Run your first automated article

This is where it all comes together.

Click on the Coordinator agent card to open the chat window.

Type something like this:

I want you to create a new article about the recent AI trends – all on your own.
Get some images, and then have the article written and wait for publishing.

Press send. The coordinator analyzes your request, divides it into subtasks and delegates it to the appropriate employees.

Keep a close eye on the response. You see the coordinator:

1. Call the Image Agent to search Pixabay for AI-related images
2. Call the WordPress Agent to write and save the article
3. Report back with a summary of what was created

The whole thing happens in less than two minutes.

Verify the result in WordPress

Go to your WordPress admin and check the post list.

Open the draft to see the full article.

The article contains properly formatted sections, embedded images with credit lines, and a logical content structure. It’s saved as a draft, so nothing goes live until you view it and click Publish.

That’s the whole point of this system. Agents do the hard work. You retain editorial control.

To proceed

How the Multi-Agent Architecture Works

The communication flow follows a simple hierarchy:

      User Request
           ↓
   Coordinator Agent
           ↓
   ┌─────────────────┐
   ↓                 ↓
Image Agent    WordPress Agent
(Pixabay MCP)  (WordPress MCP)

You speak to the coordinator. It delegates to employees. Employees use their MCP tools to perform subtasks. The results flow back to the coordinator, who prepares a final answer for you.

This pattern scales. You can add more employee agents for SEO analysis, social media posting, email newsletters, or translations. If you’re already familiar with WordPress automation plugins, consider this the next evolution: agents that don’t just follow rules, but reason about tasks and delegate work. The coordinator takes care of the routing.

Customize agent prompts

The system questions in this tutorial are intentionally minimal. For production use, you’ll want to be more specific.

Here is an example of a more detailed coordinator prompt:

You are a content publishing coordinator. When asked to create content:
1. Always search for 4-6 relevant images first
2. Ensure images have proper attribution
3. Create articles with SEO best practices
4. Set status to Draft for human review
5. Notify the user when complete

The more specific your directions, the more consistent the output.

Add other integrations

The same pattern works for any service with an MCP server or API. You can add LinkedIn, Brevo for email, or custom tools built with Apteva’s SDK. If you look at other approaches to AI-powered automation within WordPress itself, our AI Workflow Automation review covers a plugin that takes a different (but complementary) approach to the same problem.

Troubleshooting

Here you will find the most common problems and how to solve them.

MCP server not connecting: Double-check your login details. For WordPress, make sure the Stifli Flex MCP plugin is activated and your application password is valid. For Pixabay, verify your API key at pixabay.com/api/docs.

Agent shows “stopped” instead of “active”: Click the Start button on the agent card. If it still doesn’t start, check that Multi-Agent mode is enabled and that the correct MCP servers are assigned.

Images that do not appear in the article: Confirm that the Image Agent has assigned the Pixabay MCP server and that Vision capabilities are enabled. Also check that your Pixabay API quota has not been exceeded.

WordPress concept not created: Check the WordPress MCP authentication. The most common problem is an expired or revoked application password. If necessary, create a new profile based on your WordPress profile.

What’s next?

You’ve just built a multi-agent system that takes a single natural language command and turns it into a researched, illustrated WordPress concept. The coordinator delegates, the Image Agent provides images, and the WordPress Agent creates and publishes content.

The real strength here is modularity. Need an SEO agent? Create an employee with access to an SEO tool’s MCP server. Do you want automated translations? Same pattern. The architecture grows with your needs, without you having to rebuild anything. For a broader look at how AI is reshaping WordPress workflows, check out our overview of the best AI tools for WordPress web development.

I’m curious to see how people will take this further. What is the first workflow you would automate with such a configuration?