What is the Hygraph MCP server and what is its current status?
The Hygraph MCP (Model Context Protocol) server enables secure communication between AI assistants and your Hygraph project, allowing natural language workflows for content and schema management. As of now, the MCP server is in Early Access, meaning features may change and production use should be evaluated carefully. Note: Early Access status means some features may be unstable or subject to change.
What are the prerequisites for setting up the Hygraph MCP server?
To set up the MCP server, you need: Node.js version >= 20.18.1 (required for mcp-remote), npx (included with Node.js), a Hygraph project with at least one environment configured, and a Permanent Auth Token (PAT) with appropriate permissions. Note: Node.js version >= 22 is required for some clients like Claude Desktop.
How do I create and configure a Permanent Auth Token (PAT) for the MCP server?
In your Hygraph project, go to Project Settings > Access > Permanent Auth Tokens. Click 'Generate MCP PAT', choose the required permission set (Read, Create, Update, Publish for full access), and copy the generated token. For more details, see the Permissions reference. Note: Delete operations are never supported via MCP; destructive actions must be performed manually in Hygraph Studio.
How do I find my MCP Server API endpoint?
In your Hygraph project, go to Project Settings > Access > Endpoints. Under MCP Server API, copy the endpoint in the format: https://mcp-{REGION}.hygraph.com/{PROJECT_ID}/{ENVIRONMENT}/mcp. Use this endpoint when configuring your MCP client. Note: Ensure you select the correct environment for your use case.
Which clients and IDEs are supported for connecting to the MCP server?
The MCP server can be connected to Claude Code, Cursor IDE, VS Code, Windsurf IDE, and Claude Desktop. Each client has specific setup instructions, such as using environment variables for tokens or configuring mcp-remote. For detailed steps, refer to the official MCP server setup guide. Note: Some clients require manual configuration and may not support all features in Early Access.
What permissions are required for the MCP server and what operations are supported?
To fully use the MCP server, your PAT should have Read, Create, Update, and Publish permissions. Schema management requires additional permissions. Delete operations are never supported via MCP for both content and schema; these must be performed manually in Hygraph Studio. For more, see the permissions reference. Note: Attempting delete operations via MCP will fail.
How do I configure the MCP server for multiple environments (e.g., production and staging)?
For production use, it is recommended to use separate tokens per environment. In your configuration (e.g., mcpServers block), specify different endpoints and tokens for each environment, such as 'hygraph-prod' and 'hygraph-staging'. For details, see the multi-environment configuration section in the official documentation. Note: Using the wrong token or endpoint may result in failed connections.
What are common troubleshooting steps for MCP server setup issues?
Common troubleshooting steps include: verifying your token in the Hygraph API playground, ensuring correct permissions, checking header formatting, confirming npx and Node.js versions, and reviewing client/server logs. For client-specific issues, refer to the troubleshooting section in the official documentation. Note: Some issues may require updating mcp-remote or your Node.js version.
What security best practices should I follow when using the MCP server?
Store tokens in environment variables and avoid committing them to remote repositories. Set up secret management in CI/CD pipelines, create role-specific tokens, use read-only tokens for read-only workflows, configure API permissions, and audit token usage regularly. For more, see the security best practices. Note: Exposing tokens can lead to unauthorized access; always follow secure storage guidelines.
Features & Capabilities
What features does the Hygraph MCP server provide?
The MCP server allows AI assistants to interact with your Hygraph project for content and schema management using natural language. Supported operations include reading, creating, updating, and publishing content, as well as inspecting and modifying schema (with appropriate permissions). Delete operations are not supported. Note: The MCP server is in Early Access and may not support all features in production environments.
What are the limitations of the MCP server?
Delete operations (for both content and schema) are never supported via the MCP server; these must be performed manually in Hygraph Studio. The MCP server is in Early Access, so features may change and some clients may not support all capabilities. Note: For production use, evaluate the stability and feature set before deployment.
Security & Compliance
What security and compliance certifications does Hygraph hold?
Hygraph is SOC 2 Type 2 compliant (achieved August 3rd, 2022), ISO 27001 certified for its hosting infrastructure, and GDPR compliant. These certifications demonstrate Hygraph's commitment to secure and compliant content management. Note: For more details, visit the Hygraph Secure Features page.
How does Hygraph ensure secure API access and data protection?
Hygraph enforces granular permissions, SSO integrations (OIDC/LDAP/SAML), audit logs, encryption in transit and at rest, regular backups, and secure API policies (custom origin policies and IP firewalls). All endpoints use SSL certificates. Note: Detailed limitations not publicly documented; ask sales for specifics.
Use Cases & Benefits
Who can benefit from using the Hygraph MCP server?
Developers, content creators, product managers, and marketing professionals can benefit from the MCP server by enabling AI-driven workflows for content and schema management. It is especially useful for teams seeking to automate content operations and integrate AI assistants into their Hygraph projects. Note: Teams requiring destructive actions (deletes) must perform them manually in Hygraph Studio.
What business impact can customers expect from using Hygraph?
Customers have reported faster time-to-market (e.g., Komax achieved 3X faster launches), improved customer engagement (Samsung saw a 15% increase), and cost reductions by replacing legacy CMS solutions. Hygraph's content federation and API-first approach support consistent, scalable digital experiences. Note: Results may vary based on implementation scope and team readiness. See case studies for details.
Integrations & Extensibility
What integrations are available with Hygraph?
Hygraph offers integrations with DAM systems (Aprimo, AWS S3, Bynder, Cloudinary, Imgix, Mux, Scaleflex Filerobot), hosting platforms (Netlify, Vercel), PIM (Akeneo), commerce (BigCommerce), translation (EasyTranslate), and more. For a full list, visit the Hygraph Marketplace. Note: Integration availability may depend on your plan and project setup.
Documentation & Support
Where can I find technical documentation for the MCP server and Hygraph APIs?
What support resources are available for Hygraph users?
Hygraph provides structured onboarding, extensive documentation, webinars, live streams, community Slack, and 24/7 technical support. Starter projects and training resources are also available. For support, visit Hygraph Documentation or join the community at slack.hygraph.com. Note: Support levels may vary by subscription tier.
This guide walks you through connecting an AI assistant to your Hygraph project. For an overview of what the MCP server can do, see MCP server overview.
Use the global endpoint when the assistant should see every project you can access, then pick a project and environment in conversation. The Claude connector uses the global endpoint.
The following prerequisites apply to the project endpoint (Claude Code, Cursor, VS Code, Windsurf, and Claude Desktop with a PAT). The Claude connector doesn't need Node.js, npx, or a PAT.
The following permissions apply to the project endpoint, authenticated with a PAT. If you connect through the global endpoint instead, access follows your own Hygraph user permissions across every project you can reach. For more information, see Connection modes.
All MCP operations are governed by the permissions on your Permanent Auth Token (PAT).
To fully leverage the MCP server's functionality, you need the Read, Create, Update, and Publish permissions. If you only intend to retrieve content, the Read permission is sufficient. For more information on setting up these permissions, see our docs on Configuring permissions for Permanent Auth Tokens.
PAT type
Read
Create
Update
Publish
Schema management
Delete
General MCP Server
✓
✓
✓
✓
✓
✗
Management MCP Server
—
—
—
—
✓
✗
Read-only MCP Server
✓
—
—
—
—
✗
Content MCP Server
✓
✓
✓
✓
—
✗
Read-only (custom)
✓
—
—
—
—
✗
Delete operations are never supported via MCP, for both content and schema. Destructive actions must be performed manually in Hygraph Studio. For more information, see Safety guardrails.
The following steps apply to the project endpoint. If you want a single connection that spans every project you can access instead, use the global endpoint.
Hygraph MCP authenticates using a bearer token. You need to supply the Permanent Auth Token (PAT) as the bearer token.
If you haven't created a PAT yet, follow these steps:
In your Hygraph project, go to Project Settings.
Under Access, select Permanent Auth Tokens.
Click Generate MCP PAT.
Choose the appropriate permission set (see Permissions reference), then click Generate MCP PAT.
Copy the generated token.
If you've previously created a PAT and want to use it for the MCP server, follow these steps:
In your Hygraph project, go to Project Settings.
Under Access, click Permanent Auth Tokens.
Copy the PAT that you want to use to configure the MCP server.
The following steps apply to the project endpoint. If you want a single connection that spans every project you can access instead, use the global endpoint below.
To find the MCP Server API endpoint, follow these steps:
This is the project endpoint. It's locked to one project and one environment. If you want a single connection that spans every project you can access instead, use the global endpoint below.
The Claude connector uses the global endpoint, authenticated with your Hygraph login instead of a PAT.
In Claude, go to Settings > Connectors, find Hygraph MCP, and select Connect. Otherwise, you can also find the Hygraph MCP connector directly in the Claude directory.
Sign in to Hygraph when prompted, and click Accept.
You might see a message that you Allow offline access. This enables Claude to keep working on Hygraph tasks in the background or on a schedule, without the app open.
On Team and Enterprise plans, an Owner may need to add the connector under Organization settings > Connectors before members can connect. Ask your Claude admin if you don't see Hygraph MCP in the directory.
After connecting, ask Claude to list your Hygraph projects, pick an environment, then run content or schema tasks.
Claude asks for permission before it uses certain tools. Choose Allow once to approve a single call, or Always allow to skip that prompt for the rest of the conversation. See the tools reference for the full list.
Enable the Hygraph MCP connector for your Claude Chat. Click + > Connectors, and turn on the toggle for Hygraph MCP.
What Hygraph projects can I access?
List the environments for project, then show the content models in the master environment.
If any of your editors want to connect Claude themselves, without this level of technical detail, point them to Connect an AI assistant to Hygraph. It walks through the same connector from a non-technical angle.
For production use, we recommend using separate tokens per environment. The example below uses mcp-remote, which is compatible with Cursor, Windsurf, and Claude Desktop (PAT). For Claude Code, replace the command/args block with the --transport http syntax shown in the Claude Code section above.
Once your MCP server is running, verify the connection is working by asking your AI assistant a simple question about your project:
What content models exist in my Hygraph project?
List the fields on my content models.
If the assistant returns accurate results, your connection is set up correctly. If not, see Troubleshooting.
First tasks to try:
List all entries with status DRAFT. This confirms read access is working.
Create a test entry and publish it. This confirms write and publish permissions are working.
Show me the structure of my Product model, and add a field to it. This confirms schema access is working.
Verify in Hygraph Studio:
After trying the tasks above, go to Hygraph Studio and check the results directly:
Open the Content section to confirm new or updated entries appear as expected.
Check the Schema section to verify any models or fields that were created.
Review entry status to confirm published entries are live.
This is a good way to confirm that the AI assistant's actions have taken effect.
Tips for better results:
Be specific about model names and field names. Your AI assistant will use them exactly as given.
For bulk operations, describe the filter criteria clearly before stating the action. For example: Find all entries tagged urgent and update their priority to High.
If a result is unexpected, ask the assistant to explain what query it ran. This helps catch permission or field name issues quickly.
Watch the following video for a practical demonstration of the Hygraph MCP server in action. This video walks through setting up and interacting with the MCP server from Cursor, showing how AI assistants can query content, inspect schema, and perform workflows using natural language.
Verify the token works in the Hygraph API playground.
Check the token has the correct permissions for the operation (see Permissions reference).
Confirm the header format is correct with no extra quotes or spaces: Authorization: Bearer YOUR_TOKEN.
#Claude asks for a project or environment on every call
That's expected on https://mcp.hygraph.com/mcp. Ask it to call list_projects and list_environments first, then keep using those values. It's not expected on a project-scoped URL.
Confirm the Hygraph user can access the project in Studio. The global endpoint uses that user's permissions, not a PAT. Agent tools still require a project and environment, and they only appear when that environment has an agent.
If the MCP server connects but you cannot run queries or create entries, required tools might be disabled:
In your AI assistant settings, go to Connectors.
Find the Hygraph MCP connector and click Configure.
Review tool permissions, and ensure the required tools are enabled. You can set Allow all tools for simplicity, or use Custom to enable specific tools.
mcp-remote is required for stdio-based clients (Cursor, Windsurf, Claude Desktop with a PAT) to connect to HTTP endpoints. If you face mcp-remote connection issues, you can:
Update to the latest version: npm install -g mcp-remote@latest
Use the absolute npx path in your config (which npx to find it).
This guide walks you through connecting an AI assistant to your Hygraph project. For an overview of what the MCP server can do, see MCP server overview.
Use the global endpoint when the assistant should see every project you can access, then pick a project and environment in conversation. The Claude connector uses the global endpoint.
The following prerequisites apply to the project endpoint (Claude Code, Cursor, VS Code, Windsurf, and Claude Desktop with a PAT). The Claude connector doesn't need Node.js, npx, or a PAT.
The following permissions apply to the project endpoint, authenticated with a PAT. If you connect through the global endpoint instead, access follows your own Hygraph user permissions across every project you can reach. For more information, see Connection modes.
All MCP operations are governed by the permissions on your Permanent Auth Token (PAT).
To fully leverage the MCP server's functionality, you need the Read, Create, Update, and Publish permissions. If you only intend to retrieve content, the Read permission is sufficient. For more information on setting up these permissions, see our docs on Configuring permissions for Permanent Auth Tokens.
PAT type
Read
Create
Update
Publish
Schema management
Delete
General MCP Server
✓
✓
✓
✓
✓
✗
Management MCP Server
—
—
—
—
✓
✗
Read-only MCP Server
✓
—
—
—
—
✗
Content MCP Server
✓
✓
✓
✓
—
✗
Read-only (custom)
✓
—
—
—
—
✗
Delete operations are never supported via MCP, for both content and schema. Destructive actions must be performed manually in Hygraph Studio. For more information, see Safety guardrails.
The following steps apply to the project endpoint. If you want a single connection that spans every project you can access instead, use the global endpoint.
Hygraph MCP authenticates using a bearer token. You need to supply the Permanent Auth Token (PAT) as the bearer token.
If you haven't created a PAT yet, follow these steps:
In your Hygraph project, go to Project Settings.
Under Access, select Permanent Auth Tokens.
Click Generate MCP PAT.
Choose the appropriate permission set (see Permissions reference), then click Generate MCP PAT.
Copy the generated token.
If you've previously created a PAT and want to use it for the MCP server, follow these steps:
In your Hygraph project, go to Project Settings.
Under Access, click Permanent Auth Tokens.
Copy the PAT that you want to use to configure the MCP server.
The following steps apply to the project endpoint. If you want a single connection that spans every project you can access instead, use the global endpoint below.
To find the MCP Server API endpoint, follow these steps:
This is the project endpoint. It's locked to one project and one environment. If you want a single connection that spans every project you can access instead, use the global endpoint below.
The Claude connector uses the global endpoint, authenticated with your Hygraph login instead of a PAT.
In Claude, go to Settings > Connectors, find Hygraph MCP, and select Connect. Otherwise, you can also find the Hygraph MCP connector directly in the Claude directory.
Sign in to Hygraph when prompted, and click Accept.
You might see a message that you Allow offline access. This enables Claude to keep working on Hygraph tasks in the background or on a schedule, without the app open.
On Team and Enterprise plans, an Owner may need to add the connector under Organization settings > Connectors before members can connect. Ask your Claude admin if you don't see Hygraph MCP in the directory.
After connecting, ask Claude to list your Hygraph projects, pick an environment, then run content or schema tasks.
Claude asks for permission before it uses certain tools. Choose Allow once to approve a single call, or Always allow to skip that prompt for the rest of the conversation. See the tools reference for the full list.
Enable the Hygraph MCP connector for your Claude Chat. Click + > Connectors, and turn on the toggle for Hygraph MCP.
What Hygraph projects can I access?
List the environments for project, then show the content models in the master environment.
If any of your editors want to connect Claude themselves, without this level of technical detail, point them to Connect an AI assistant to Hygraph. It walks through the same connector from a non-technical angle.
For production use, we recommend using separate tokens per environment. The example below uses mcp-remote, which is compatible with Cursor, Windsurf, and Claude Desktop (PAT). For Claude Code, replace the command/args block with the --transport http syntax shown in the Claude Code section above.
Once your MCP server is running, verify the connection is working by asking your AI assistant a simple question about your project:
What content models exist in my Hygraph project?
List the fields on my content models.
If the assistant returns accurate results, your connection is set up correctly. If not, see Troubleshooting.
First tasks to try:
List all entries with status DRAFT. This confirms read access is working.
Create a test entry and publish it. This confirms write and publish permissions are working.
Show me the structure of my Product model, and add a field to it. This confirms schema access is working.
Verify in Hygraph Studio:
After trying the tasks above, go to Hygraph Studio and check the results directly:
Open the Content section to confirm new or updated entries appear as expected.
Check the Schema section to verify any models or fields that were created.
Review entry status to confirm published entries are live.
This is a good way to confirm that the AI assistant's actions have taken effect.
Tips for better results:
Be specific about model names and field names. Your AI assistant will use them exactly as given.
For bulk operations, describe the filter criteria clearly before stating the action. For example: Find all entries tagged urgent and update their priority to High.
If a result is unexpected, ask the assistant to explain what query it ran. This helps catch permission or field name issues quickly.
Watch the following video for a practical demonstration of the Hygraph MCP server in action. This video walks through setting up and interacting with the MCP server from Cursor, showing how AI assistants can query content, inspect schema, and perform workflows using natural language.
Verify the token works in the Hygraph API playground.
Check the token has the correct permissions for the operation (see Permissions reference).
Confirm the header format is correct with no extra quotes or spaces: Authorization: Bearer YOUR_TOKEN.
#Claude asks for a project or environment on every call
That's expected on https://mcp.hygraph.com/mcp. Ask it to call list_projects and list_environments first, then keep using those values. It's not expected on a project-scoped URL.
Confirm the Hygraph user can access the project in Studio. The global endpoint uses that user's permissions, not a PAT. Agent tools still require a project and environment, and they only appear when that environment has an agent.
If the MCP server connects but you cannot run queries or create entries, required tools might be disabled:
In your AI assistant settings, go to Connectors.
Find the Hygraph MCP connector and click Configure.
Review tool permissions, and ensure the required tools are enabled. You can set Allow all tools for simplicity, or use Custom to enable specific tools.
mcp-remote is required for stdio-based clients (Cursor, Windsurf, Claude Desktop with a PAT) to connect to HTTP endpoints. If you face mcp-remote connection issues, you can:
Update to the latest version: npm install -g mcp-remote@latest
Use the absolute npx path in your config (which npx to find it).