How to Schedule Social Media Posts from Claude and Cursor Using posterly's MCP Server

How to Schedule Social Media Posts from Claude and Cursor Using posterly's MCP Server

If you already spend time inside AI tools like Claude or Cursor, there is now a faster way to manage your social media. Instead of switching between tabs, you can schedule posts, check your accounts, and find the next open time slot without ever leaving your AI assistant.

posterly's MCP (Model Context Protocol) server makes this possible. It takes about five minutes to set up and works with any MCP-compatible client.

Here is what you can do once it is connected:

• Schedule posts to any of your connected social accounts using plain English
• Upload media and attach it to a post in the same conversation
• List accounts to see which platforms are ready to publish to
• Find the next available slot so your posts never pile up at the same time

This guide covers the full setup process, the available tools, and some practical workflows that save real time.

Why Use the posterly MCP Server?

Most social media scheduling tools require you to log into a dashboard, pick a platform, write a caption, choose a time, and click publish. That workflow is fine when you are batch-scheduling content. But if you are already in Claude writing copy, or in Cursor working on a product launch, jumping into a separate tool breaks your flow.

The MCP server turns posterly into a tool your AI assistant can use directly. You describe what you want to post, and the assistant handles the rest. No tab switching. No copy-pasting captions from one window to another.

It also opens up automation possibilities for developers. Because the underlying REST API powers the MCP server, anything the server can do can also be built into your own apps, scripts, or internal tools. You can find the full API documentation at poster.ly/docs/integrations/api.

What You Will Need

Before starting, make sure you have:

• A posterly account (sign up at poster.ly with a free 7-day trial)
• At least one social account connected in posterly
• A posterly API key (found in Dashboard > Settings > API Keys)
• Node.js 18 or later installed on your machine
• An MCP-compatible AI client: Claude Desktop, Cursor, or similar

Step 1: Install the MCP Server

Open your terminal and run:

npm install -g posterly-mcp-server

This installs the server globally so it is available as a command from any directory.

Step 2: Add posterly to Your MCP Client Configuration

The exact config file location depends on which client you are using.

For Claude Desktop

Open your Claude Desktop config file. On macOS it lives at ~/Library/Application Support/Claude/claude_desktop_config.json. Add the following inside the mcpServers object:

{
  "mcpServers": {
    "posterly": {
      "command": "posterly-mcp-server",
      "env": {
        "POSTERLY_API_KEY": "your-api-key-here"
      }
    }
  }
}

For Cursor

Open Cursor's MCP settings (Settings > MCP) and add the same block. Cursor will detect and load the server when you restart it.

For Other MCP Clients

Any client that supports the Model Context Protocol will work. The configuration format is the same. Replace your-api-key-here with the key from your posterly dashboard and you are done.

Step 3: Restart Your AI Client

After saving the config, restart Claude Desktop or Cursor fully. The posterly tools will appear in your assistant's available tool list. You can verify this by asking: "What posterly tools do you have access to?"

The Four Tools Available

Once connected, your AI assistant has access to these posterly tools:

Tool	What It Does
list_accounts	Returns all connected social accounts with their platform and username
create_post	Schedules a post with caption, platform, account, and time
upload_media	Uploads an image or video and returns a media ID for use in posts
find_next_slot	Finds the next available posting time for a given account

These four tools cover the core scheduling workflow. You can string them together in a single conversation to go from idea to scheduled post in under a minute.

Practical Workflows

Scheduling a Simple Post

In Claude or Cursor, you can say:

"Schedule a LinkedIn post for tomorrow at 10am announcing that we just launched our new pricing page. Keep it professional and under 200 words."

The assistant will use list_accounts to find your LinkedIn account, write the caption, and call create_post to schedule it. You get a confirmation with the scheduled time.

Posting With an Image

If you have an image file or a publicly accessible image URL, you can ask:

"Upload this image and schedule an Instagram post with it for Friday at 12pm. Caption should highlight our spring collection."

The assistant calls upload_media first to get a media ID, then passes it to create_post along with the caption and schedule.

Finding a Gap in Your Calendar

If you are not sure when to post next, just ask:

"When is the next available posting slot for my Instagram account?"

The assistant calls find_next_slot and returns the first open time, respecting the 1-hour gap between posts that posterly's smart scheduler enforces by default.

Running a Multi-Platform Campaign From One Conversation

One of the most powerful workflows is running an entire launch campaign from a single chat:

1. Ask the assistant to list your accounts
2. Describe your campaign: platforms, messaging, timing
3. The assistant drafts and schedules each post, one by one, confirming each before proceeding

This works especially well inside Claude, where the conversation context lets the assistant keep track of what has already been scheduled and what still needs to go out.

Using the REST API for Custom Integrations

The MCP server is built on top of posterly's REST API. If you want to build your own integration, such as a Zapier-style automation, an internal dashboard, or a GitHub action that posts release notes automatically, the API gives you full access to the same tools.

You can generate an API key from your posterly account settings. The API uses standard bearer token authentication and returns JSON responses.

Some teams use the API to:

• Post product update announcements when a new GitHub release is tagged
• Schedule customer success posts from a CRM trigger
• Push scheduled content from a headless CMS directly into the posterly queue

The Ship and Share feature is a prebuilt version of this for GitHub commits, if you want something that works without any custom code.

Tips for Getting the Most Out of the MCP Integration

Be specific about timing. The more specific you are, the less back-and-forth happens. Instead of "schedule this for later", say "schedule this for Wednesday at 9am Pacific".

Use account handles in your requests. If you have multiple accounts on the same platform, referring to them by handle avoids ambiguity. For example: "Post this to my @brandaccount Instagram, not my personal one."

Keep your API key out of shared files. The API key gives full access to your posterly account. Store it in an environment variable or a secrets manager rather than pasting it directly into any config file that might be committed to version control.

Check your credit balance before large campaigns. AI caption generation and image uploads use posterly credits. You can check your balance in Dashboard > Account.

Frequently Asked Questions

Which AI assistants support the posterly MCP server?

Any AI assistant that supports the Model Context Protocol works with the posterly MCP server. This includes Claude Desktop, Cursor, and other MCP-compatible clients. The setup process is the same for all of them: install the npm package, add the config block, and restart the client.

Is the MCP server the same as the REST API?

They use the same underlying API, but they serve different use cases. The MCP server is designed for use inside AI assistants, so your AI can call posterly tools on your behalf during a conversation. The REST API is for developers who want to build their own integrations directly. Both require an API key.

Do I need a paid plan to use the API or MCP server?

API access is available on all posterly plans including the Starter tier. Some advanced features may require higher plan tiers. You can check what is included on each plan at poster.ly.

Can I use the MCP server to schedule posts to all 9 platforms?

Yes. The MCP server works with all platforms you have connected in posterly, including LinkedIn, Instagram, X, Facebook, TikTok, YouTube, Threads, Pinterest, and Google Business Profile. The list_accounts tool will show you exactly which accounts are available.

What happens if the AI assistant makes a mistake, like scheduling to the wrong account?

Posts scheduled via the MCP server appear in your posterly calendar just like any other post. You can review, edit, or delete them before they go live. Nothing publishes without passing through the standard posterly queue.