Frequently Asked Questions

Product Overview & Use Cases

What is the Hygraph MCP server and what does it do?

The Hygraph MCP server is an Early Access feature that connects AI assistants (such as Claude, Cursor, Windsurf, and VS Code) to your Hygraph project using the Model Context Protocol (MCP). It provides a permission-aware, standardized interface to Hygraph's APIs, allowing AI assistants to perform content and schema operations using natural language prompts. Instead of manual setup, you can describe tasks in plain language, enabling actions like bulk content updates, schema creation, and cross-environment migrations. Note: As an Early Access feature, production use should be evaluated carefully and features may change. Learn more.

What types of tasks can AI assistants perform using the MCP server?

With the appropriate permissions, AI assistants can read, filter, create, update, and publish content entries, perform bulk updates, and move content between environments. For schema operations, assistants can inspect models, retrieve model structures, create new models, add or update fields, manage enumerations and components, and move schema changes between environments. Destructive actions like delete and unpublish are intentionally not supported via MCP for safety. Note: Schema and content operations are governed by the permissions on your Permanent Auth Token (PAT). See details.

Which AI assistants and tools are compatible with the Hygraph MCP server?

The Hygraph MCP server supports integration with AI assistants and tools that implement the Model Context Protocol (MCP), including Claude, Cursor, Windsurf, and VS Code. This allows you to connect the same Hygraph project to multiple MCP-compatible tools using a consistent server interface. Note: Compatibility depends on the assistant's support for MCP and your project configuration. Learn more about MCP.

Features & Capabilities

What are the key benefits of using the Hygraph MCP server?

Key benefits include:

Note: As an Early Access feature, some capabilities may change and not all production scenarios are covered. See full list.

What operations are intentionally restricted or not supported by the MCP server?

The MCP server does not support destructive actions such as deleting entries or schema elements, or unpublishing entries. These operations must be performed manually in Hygraph Studio to prevent accidental data loss in automated workflows. Note: This is a safety guardrail and may limit certain advanced automation scenarios. Read more.

How does the MCP server handle permissions and security for AI assistants?

All MCP operations are governed by the permissions set on your Permanent Auth Token (PAT). You can choose from different PAT types (General, Management, Content, or custom Read-only) to control which actions an AI assistant can perform. For example, delete is never allowed via MCP, and schema management is only available to certain PAT types. Note: Misconfigured tokens may result in insufficient or excessive permissions; review the permissions reference for details.

Can you provide examples of how to use natural language prompts with the MCP server?

Yes. Example prompts include:

These prompts allow AI assistants to automate content and schema operations without manual scripting. Note: Complex or custom workflows may require manual intervention or additional configuration. See more examples.

Technical Requirements & Implementation

What are the requirements for setting up and using the MCP server?

To use the MCP server, you need a Hygraph project, an MCP-compatible AI assistant or tool, and a properly configured Permanent Auth Token (PAT) with the necessary permissions. Separate endpoints and tokens are recommended for different environments (e.g., staging vs production) to support safe deployment and review. For setup instructions, see the MCP server setup guide. Note: As MCP server is in Early Access, some setup steps or requirements may change.

Where can I find technical documentation and setup guides for the MCP server?

Comprehensive technical documentation, including setup instructions, permissions reference, and integration guides for the MCP server, is available at https://hygraph.com/docs/hygraph-ai/mcp-server and https://hygraph.com/docs/hygraph-ai/mcp-server-setup. For broader API and integration documentation, visit Hygraph API Reference. Note: Documentation is updated as features evolve during Early Access.

Security & Compliance

What security and compliance certifications does Hygraph hold?

Hygraph is SOC 2 Type 2 compliant (as of August 3rd, 2022), ISO 27001 certified for hosting infrastructure, and GDPR compliant. These certifications ensure that the platform meets international standards for information security and data protection. For more details, visit the Hygraph Secure Features page. Note: While MCP server operations are permission-aware, always review your token and environment configurations for compliance with your organization's policies.

Limitations & Early Access Considerations

Are there any limitations or risks to using the MCP server in production?

The MCP server is currently in Early Access, meaning features may change and not all production scenarios are fully supported. Destructive actions (delete, unpublish) are intentionally blocked to reduce operational risk, but this may limit advanced automation. Evaluate production use carefully and monitor updates to documentation and feature set. For critical workflows, manual review and fallback to Hygraph Studio may be necessary. See Early Access notes.

LLM optimization

When was this page last updated?

This page wast last updated on 12/12/2025 .

Hygraph
Docs

#MCP server

The Hygraph MCP server connects AI assistants to your Hygraph project using the Model Context Protocol (MCP). Connect the same project from Claude (Connector or Claude Code), Cursor, VS Code, or Windsurf, through a standardized interface to Hygraph's APIs.

Instead of setting up everything manually in Studio, you describe what you want in plain language. Tasks that previously required a developer, such as bulk content updates, schema creation, and cross-environment migrations, can now be done with a single prompt.

#Key benefits

  • Permission-aware by default: MCP operations follow the permissions on your Permanent Auth Token (PAT), or on your own Hygraph user account when you connect through the global endpoint, so assistants can only perform actions you explicitly allow.
  • Faster workflow automation: Use natural language to query, create, and publish content or apply structured bulk updates without writing one-off scripts for each task.
  • Standardized integration across clients: Connect the same Hygraph project to multiple MCP-compatible tools, such as Claude, Cursor, VS Code, and Windsurf, using a consistent server interface.
  • Reduced operational risk: Destructive actions like delete and unpublish are not supported via MCP, helping prevent accidental data loss in automated workflows.
  • Environment-aware workflows: Use separate endpoints and tokens per environment, such as staging vs production, to support safer deployment and review processes. You can connect to the MCP server via a project-scoped endpoint with a Permanent Auth Token, or the global endpoint at https://mcp.hygraph.com/mcp with your Hygraph account, so the assistant can list every project and environment you have access to.

#What you can do

#Content operations

With the appropriate permissions (for example, a General MCP Server or Content MCP Server PAT), AI assistants can:

  • Read entries
  • Filter and query entries
  • Create new entries
  • Update existing entries
  • Publish entries
  • Perform bulk updates
  • Move content between environments

#Schema operations

With the correct permissions (for example, a General MCP Server or Management MCP Server PAT), AI assistants can:

  • Inspect existing content models
  • Retrieve the structure of a specific model (fields, types, validations)
  • Create new content models
  • Add new fields to existing models
  • Update field configurations
  • Manage enumerations and components
  • Move schema changes between environments
  • Perform structured schema refactoring tasks

#Agent operations

An agent saved in Studio with Use in Agent runs is available to use with MCP. These agents are available from the content table and the entry form. Agents saved with Use in workflows still run only when an entry reaches that workflow step in Studio; MCP cannot start those.

With an MCP connection to an environment that has at least one agent, AI assistants can:

  • List the agents configured in that environment
  • Trigger the agent on one or more entries
  • Check run status for those entries

Agent runs from MCP follow the same rules as manual agent runs in Studio: changes land in DRAFT and wait for editorial review. MCP does not publish agent output.

#AI guidelines

With guidelines configured in the project, you can load brand voice, glossary, audience, and compliance context before creating or updating entries. This is read-only. It doesn't create, edit, or assign guidelines. See AI Guidelines for more information.

#Availability

  • All Hygraph projects

#Permissions

All MCP operations are governed by the permissions on your Permanent Auth Token (PAT), or by your own Hygraph user permissions when you connect through the global endpoint. Use this table to choose the right token type for your workflow.

PAT typeReadCreateUpdatePublishSchema managementDelete
General MCP Server✓✓✓✓✓✗
Management MCP Server————✓✗
Content MCP Server✓✓✓✓—✗
Read-only (custom)✓————✗

For full details on creating and configuring tokens, see MCP server setup.

#Example prompts

The following example prompts demonstrate how you can use natural language to interact with your Hygraph project through the MCP server.

Content queries:

  • List all blog posts in Hygraph published this week.
  • Find all entries in Hygraph with status DRAFT that were last updated more than 90 days ago.

Content updates:

  • Find all posts in Hygraph with status DRAFT and update their priority to High.
  • Publish all entries in Hygraph currently in review.

Multi-step workflows:

  • Find all posts in Hygraph with urgent in the title, update their priority to High, then publish them.
  • Move all published entries in Hygraph to the staging environment.

Schema operations:

  • Show me the structure of the Product model in Hygraph.
  • Create a new model in Hygraph called Author with name, bio, and avatar fields.
  • Add a required SEO Title field to all models in Hygraph that represent pages.
  • Add a new enumeration in Hygraph called Priority with values Low, Medium, High.

Agent runs:

  • List the AI agents in this Hygraph environment and tell me which ones I can trigger from here.
  • Run the translator agent on these draft blog posts, then check progress until they are ready for review.

AI guidelines:

  • Load the content-generation guidelines for this Hygraph project, then draft a new Product entry that follows them.

Combined workflows:

  • List all blog posts from Hygraph and create individual Markdown files for each post with front matter including slug, title, and publish date.
  • Generate TypeScript types for all content models in Hygraph.

#Safety guardrails

The Hygraph MCP server is designed to prevent accidental data loss in automated workflows. The following operations are intentionally not supported via MCP and must be performed manually in Hygraph Studio:

  • Delete entries or schema elements
  • Unpublish entries

This ensures that AI-assisted workflows cannot unintentionally remove or revert production content or schema.

#What's next

  • MCP server setup: Connect your AI assistant to your Hygraph project, and choose between a project-scoped endpoint and the global endpoint.
  • MCP tools reference: Look up every MCP tool, what it does, and which endpoint it's available on.
  • AI Agents: Understand the agent types available in Studio and how manual and workflow-bound agents differ.
  • AI Guidelines: Create and assign guidelines that AI tools can read.