MCP Integration

Connect AI assistants like Claude, Cursor, and other MCP-compatible tools to Waldium. Generate, schedule, and publish AI blog posts directly from your AI workflow.

What is MCP?

Model Context Protocol (MCP) is an open standard for connecting AI applications to external systems. Think of it like a USB-C port for AI applications. With MCP, AI assistants can interact with Waldium directly, enabling you to create and manage blog content without leaving your AI workflow.

Quick Start

Step 1: Get your MCP server URL

Your Waldium MCP server URL is:

https://mcp.waldium.com/mcp/waldium-mcp

Step 2: Configure your MCP client

Add the Waldium server to your MCP client configuration:

{
  "mcpServers": {
    "waldium": {
      "url": "https://mcp.waldium.com/mcp/waldium-mcp"
    }
  }
}

Step 3: Authenticate

When you first connect, you'll be prompted to authenticate with your Waldium account using OAuth. This securely links your AI assistant to your Waldium site.

Available Tools

The Waldium MCP server provides the following tools for AI assistants:

Use Cases

Here are common ways to use the Waldium MCP integration:

Generate Technical Documentation

"Write a technical guide explaining how to integrate our REST API 
with authentication examples and error handling"

The AI will call waldium_generate_post with your topic and parameters to create a complete blog post.

Create Engineering Blog Posts

"Write a blog post about how we migrated from PostgreSQL to 
CockroachDB, targeting senior engineers"

Knowledge-Enhanced Technical Writing

"Generate API documentation for our webhook endpoints using 
the OpenAPI spec I uploaded"

With useKnowledge: true, the AI incorporates your uploaded reference materials for more accurate, brand-aligned content.

Tool Reference

Generate Post

Generate an AI-powered blog post with full customization options.

{
  "topic": "Building REST APIs with Node.js",
  "style": "technical",
  "tone": "informative",
  "length": "medium",
  "targetAudience": "backend developers",
  "model": "gpt-5-nano",
  "isDraft": true
}

Create Post

Create a new blog post with manual content.

{
  "title": "How to Build a REST API",
  "content": "# Introduction\n\nThis guide covers...",
  "description": "A comprehensive guide to REST APIs",
  "category": "Engineering",
  "tags": ["api", "tutorial", "backend"],
  "status": "DRAFT"
}

Update Post Status

Change a post's visibility status.

{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "status": "PUBLISHED"
}

Important Notes

• Authentication: All endpoints require Bearer token authentication via API key or OAuth
• Post IDs: Post IDs are UUIDs. Validate format before operations
• Status values: Use DRAFT (hidden) or PUBLISHED (public)
• Knowledge uploads: Maximum 10MB per file. Supported formats: PDF, TXT, MD, CSV, DOC, DOCX
• Rate limiting: Respect 429 responses with exponential backoff
• AI generation time: Typically 10-30 seconds depending on length and model

Troubleshooting

Connection Issues

If you're having trouble connecting:

1. Verify your API key is valid in your dashboard
2. Check that your MCP client supports OAuth 2.0 authentication
3. Ensure you're using the correct server URL: https://mcp.waldium.com/mcp/waldium-mcp

Generation Failures

If post generation fails:

1. Check that the topic is clear and descriptive
2. Verify your account has available generation credits
3. Try a different AI model (e.g., switch from gpt-5 to gpt-5-nano)

Knowledge Base Issues

If knowledge-enhanced generation isn't working:

1. Confirm files are uploaded and processed (check Knowledge Base in dashboard)
2. Verify useKnowledge: true is set in your request
3. Check that selectedFileIds contains valid file UUIDs