mcp-server · kontent-ai — Claude Code MCP server for Kontent.ai
Hooks your Kontent.ai project into an AI editor so content types, taxonomies, assets and language versions can be handled in plain language.
claude mcp add --env KONTENT_API_KEY=<KONTENT_API_KEY> --env KONTENT_ENVIRONMENT_ID=<KONTENT_ENVIRONMENT_ID> --env PORT=<PORT> --transport stdio kontent-ai-mcp-server -- npx -y @kontent-ai/mcp-serverWhat mcp-server does
The server speaks Model Context Protocol and exposes the Kontent.ai Management API to your agent. Content types, snippets, taxonomy groups, assets and asset folders can all be listed and created. Content items get their own toolset: the container itself, language variants, translations, drafts and published versions.
Type and taxonomy edits run as patch operations — move, addInto, remove, replace. Call get-patch-guide first: it returns the allowed operations per entity type, and skipping it means the agent guesses and the request fails. Semantic search over variants exists, though it filters only by variant ID.
Two transports are available: npx @kontent-ai/mcp-server@latest stdio and npx @kontent-ai/mcp-server@latest shttp. You need a Kontent.ai account, a project, a Management API key with the right permissions, and the Environment ID. Both values go in through KONTENT_API_KEY and KONTENT_ENVIRONMENT_ID.
What it helps with
- turn a diagram into live content types
- find content by meaning when exact keywords are missing
- add a language version and set its contributor
- rename and reorganise asset folders
- check where a content type is used
Example prompts for the agent
- List every content type in the project and show where
articleis used - Create a content type
blog-postwith title and body elements - Find items about shipping and show me their drafts
How to install mcp-server
Platform CLI
claude mcp add --env KONTENT_API_KEY=<KONTENT_API_KEY> --env KONTENT_ENVIRONMENT_ID=<KONTENT_ENVIRONMENT_ID> --env PORT=<PORT> --transport stdio kontent-ai-mcp-server -- npx -y @kontent-ai/mcp-serverReplace <…> with your own values
Platform CLI
codex mcp add kontent-ai-mcp-server --env KONTENT_API_KEY=<KONTENT_API_KEY> --env KONTENT_ENVIRONMENT_ID=<KONTENT_ENVIRONMENT_ID> --env PORT=<PORT> -- npx -y @kontent-ai/mcp-serverReplace <…> with your own values
Config file
{
"mcpServers": {
"kontent-ai-mcp-server": {
"command": "npx",
"args": [
"-y",
"@kontent-ai/mcp-server"
],
"env": {
"KONTENT_API_KEY": "<KONTENT_API_KEY>",
"KONTENT_ENVIRONMENT_ID": "<KONTENT_ENVIRONMENT_ID>",
"PORT": "<PORT>"
}
}
}
}Replace <…> with your own values
Config file
{
"servers": {
"kontent-ai-mcp-server": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@kontent-ai/mcp-server"
],
"env": {
"KONTENT_API_KEY": "<KONTENT_API_KEY>",
"KONTENT_ENVIRONMENT_ID": "<KONTENT_ENVIRONMENT_ID>",
"PORT": "<PORT>"
}
}
}
}Replace <…> with your own values
Platform CLI
gemini mcp add -e KONTENT_API_KEY=<KONTENT_API_KEY> -e KONTENT_ENVIRONMENT_ID=<KONTENT_ENVIRONMENT_ID> -e PORT=<PORT> kontent-ai-mcp-server npx -y @kontent-ai/mcp-serverReplace <…> with your own values
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about mcp-server
How do I install mcp-server in Claude Code?
Run claude mcp add mcp-server -e KONTENT_API_KEY=<KONTENT_API_KEY> -e KONTENT_ENVIRONMENT_ID=<KONTENT_ENVIRONMENT_ID> -- npx -y @kontent-ai/mcp-server. Fill in your own Management API key and Environment ID.
What does mcp-server do?
It gives an AI agent tools for a Kontent.ai project: content types, snippets, taxonomies, assets, and content items with their language variants.
What do I need before running it?
A Kontent.ai account, a project, a Management API key with suitable permissions, and the Environment ID. Without the key and ID the server cannot reach the project.
Similar MCP servers
All MCP servers for Agent setup: memory, context, orchestration →mcp-server by other authors comes first.
