Documentation

MCP Server

The Clustova MCP server lets you generate SEO-optimized content directly inside your AI coding assistant — Claude Desktop, Cursor, VS Code, or Windsurf — without switching between tools.

What is MCP?

The Model Context Protocol (MCP) is an open standard that lets AI applications connect to external tools and data sources. Clustova implements an MCP server that exposes content generation as a tool your AI assistant can call.

Installation

Install the MCP server globally via npm:

bash

npm install -g clustova-mcp

Or run on-demand without installing:

bash

npx clustova-mcp

Getting Your API Token

Go to Dashboard → Sites

Click New Connection, Select MCP Option and Generate MCP Token

Copy the token and paste it into the CLUSTOVA_API_TOKEN field in your config

MCP tokens are long-lived and scoped to your account. Treat them like passwords — do not commit them to public repositories.

Configuration

Add the server to your AI client's MCP configuration:

Claude Desktop

claude_desktop_config.json

json

{
  "mcpServers": {
    "clustova": {
      "command": "npx",
      "args": ["-y", "clustova-mcp"],
      "env": {
        "CLUSTOVA_API_TOKEN": "your-api-token"
      }
    }
  }
}

Cursor

.cursor/mcp.json

json

{
  "mcpServers": {
    "clustova": {
      "command": "npx",
      "args": ["-y", "clustova-mcp"],
      "env": {
        "CLUSTOVA_API_TOKEN": "your-api-token"
      }
    }
  }
}

VS Code

.vscode/mcp.json

json

{
  "servers": {
    "Clustova": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "clustova-mcp"],
      "env": {
        "CLUSTOVA_API_TOKEN": "your-api-token"
      }
    }
  }
}

Available Tools

The Clustova MCP server exposes four tools to your AI assistant:

generate_content

Submits a content generation job and returns a jobId. Generation is asynchronous — use check_status to poll for completion.

Example prompts:

→“Write a blog post comparing Kubernetes vs Docker Swarm, 2000 words, professional tone”
→“Generate a tutorial about setting up React with TypeScript”
→“Create 3 product descriptions for wireless earbuds”

get_quota

Returns your current word balance, monthly allocation remaining, topics limit, and plan details. Useful to check before submitting a large batch.

check_status

Polls the progress of a submitted job using its jobId. Returns the current stage (SERP analysis / content generation / humanization / completed) and a completion percentage.

list_jobs

Returns a list of your recent content generation jobs with their status, topic count, model used, and creation date.

Async Workflow

Content generation is asynchronous. The typical flow is:

Submit — Ask your AI assistant to generate content → it calls generate_content and receives a jobId

Check — Ask for the status → it calls check_status with the jobId

Retrieve — When status is completed, check_status returns the full article in the response

Output Format

Every completed article is returned as markdown with:

YAML Frontmatter — title, slug, seo_title, meta_description, focus_keyword, tags, schema_type, og_description, twitter_description, word_count
Markdown Body — Clean markdown with headings, tables, code blocks, and lists
JSON-LD Schema — Article, HowTo, Product, or FAQ schema markup for search engines

Supported Parameters

Parameter	Options
Content type	blog_post, tutorial, product_description, news, review, comparison, howto, listicle, affiliate
Word count	100 – 3,000
Tone	professional, casual, friendly, authoritative
Language	30+ languages including English, Spanish, French, German, Chinese, Japanese, Arabic
Country targeting	ISO country code for localised SERP analysis (e.g., us, gb, de)

NPM Package

The Clustova MCP server is available on npm as clustova-mcp. Source code and documentation are on GitHub.

View on npm Get Started