@contentful/mcp-server

What it does

This MCP server connects Claude and other AI assistants to Contentful's Management API, enabling natural-language content operations. You can create and publish entries, edit existing content in bulk, upload and organize assets, manage content models, and automate workflows. The server enforces optional write protections on specified environments—useful for locking down production while keeping staging open for AI-assisted iteration.

Who it's for

Content managers and developers who want to integrate Contentful into AI-assisted workflows. Teams that delegate content creation, updates, and asset organization to AI while maintaining environment-level access controls.

Common use cases

• Create blog posts, product entries, or marketing content by describing requirements in natural language.
• Bulk-update entries to propagate schema changes, pricing updates, or new metadata across your content base.
• Upload and organize media assets by campaign, category, or metadata.
• Automate content localization—translate entries to multiple languages or duplicate content across regional spaces.
• Extend content models by adding fields or content types without leaving your AI assistant.

Setup pitfalls

• CONTENTFUL_MANAGEMENT_ACCESS_TOKEN and SPACE_ID are required environment variables and must be set before the server starts; the token grants full write access, so rotate it if exposed.
• The project's CI is not currently passing—test compatibility with your Node.js version before deploying to production.
• PROTECTED_ENVIRONMENTS is opt-in; without it, all environments are writable including production. Use a comma-separated list (e.g., master,staging) to block writes.
• Environment IDs in Contentful are case-sensitive; master and Master are different targets, so verify the exact ID you want to protect.