home / mcp / skolverket mcp server

Skolverket MCP Server

Provides access to Skolverket's open APIs for curriculum, school units, and planned educations to support AI-powered analysis and planning.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "isakskogstad-skolverket-mcp": {
      "url": "https://skolverket-mcp.onrender.com/mcp"
    }
  }
}

This MCP server gives AI assistants access to Skolverket’s open APIs, enabling you to search, compare, and analyze curriculum data, school units, and planned educations. You can connect to a hosted MCP server or run a local instance and use it with your preferred AI chat or coding assistant to query data, extract insights, and generate analyses.

How to use

To use the Skolverket MCP Server, connect your AI assistant to either the remote MCP endpoint or a local MCP instance. Use the MCP URL if you want a straightforward, hosted connection, or run the server locally to tailor the integration to your environment. You can query Syllabus data, School Units, and Planned Educations, and you can combine data sources to perform tasks like curriculum comparisons, program planning, and data-driven analyses.

Practical usage patterns you can perform with your AI assistant include:

How to install

Prerequisites: you need Node.js and npm installed on your system. You should have a modern terminal or shell and network access to download dependencies.

# 1) Clone the MCP server repository
git clone https://github.com/isakskogstad/skolverket-mcp.git
cd skolverket-mcp

# 2) Install dependencies
npm install

# 3) Build the project
npm run build

Run or configure the MCP server for your environment. You have two main options: use the hosted MCP URL or run a local instance that uses stdio transport.

# 4A) Use the hosted MCP URL with a client supporting HTTP transport
# The MCP URL is: https://skolverket-mcp.onrender.com/mcp

# 4B) Run a local instance and connect via stdio transport
node dist/index.js

If you run locally, ensure your client is pointed at the local index or uses stdio transport as shown in the configuration examples.

Additional configuration and notes

The MCP server supports multiple connection methods. The remote HTTP endpoint provides a simple, out-of-the-box option for AI integrations. Local setups require starting the MCP server and wiring your client to the local process using stdio transport.

Environment and transport details are demonstrated in configuration examples for popular clients. Below are representative code blocks you may adapt for your setup.

{
  "mcpServers": {
    "skolverket": {
      "type": "http",
      "url": "https://skolverket-mcp.onrender.com/mcp",
      "args": []
    }
  }
}

Security and maintenance notes

Only expose the MCP endpoint to trusted AI clients. If you run locally, keep the server and dependencies up to date with regular npm install and builds. Monitor access patterns and rotate any credentials used by clients. Maintain separate environments for development, testing, and production to minimize the risk of data leakage.

Available tools

syllabus_api

Access curriculum details, subjects, courses, and central content for various education programs.

school_units_api

Query and filter schools, preschools, and other school units, including active, closed, and dormant units.

planned_educations_api

Retrieve programs like vocational training, SFI, and adult education with start dates and enrollment data.

api_info

Provide basic information about available APIs and data sources.

curriculum_codes

Lookup codes for curricula, subjects, and courses.

prompt_templates

Prebuilt prompts for course analysis, version comparisons, adult education guidance, and course planning.