- MAPPING
- FETCHING
- VALIDATION
- HUMAN
Short answer
How can my AI see my Hygraph content models?
Your AI uses introspection to map out your content types and fields. Once it understands the schema, it writes and runs GraphQL queries to grab the exact data you need. You get structured content without writing code.
The outcomes
What you get from this connection
The AI handles the heavy lifting of navigating your headless CMS.
MAPPING
Full schema visibility
The AI reads your entire content model. It knows which models exist and what fields they contain.
FETCHING
Direct data access
Your agent runs GraphQL queries to pull specific entries or lists. It brings the content straight to your chat.
VALIDATION
Field accuracy
The AI checks field types and requirements. This ensures the queries it builds actually work against your Hygraph setup.
HUMAN
Manual oversight
You stay in control of the data. You decide which queries to run and review the results before using them.
The workflow
What your AI does when the connection is live.
The AI moves from discovery to execution in three logical steps.
Discover the schema
The AI looks at your Hygraph instance to see how your content is organized. It identifies all available models and their relationships.
list_schema_introspectionMap specific fields
Once it finds a model, the AI drills down into the specific fields. It learns the exact names and types of every attribute in that model.
get_model_fieldsRun GraphQL queries
The AI builds a query based on your request. It sends the request to Hygraph and returns the data to your chat.
execute_graphql_queryRefine data requests
If a query fails or returns the wrong data, the AI uses the schema information to fix the query and try again.
list_schema_introspectionexecute_graphql_query
Try it
Copy these to start.
Use these prompts to kick off the process.
Starting points
Replace the bracketed text with your actual Hygraph model names or field names.
Hygraph Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_E2kDf3rmUJJgcNZKXaOKQyqaD4p2MZcV1xyGfKJa/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Hygraph capabilities are ready to use.
{
"mcpServers": {
"hygraph-headless-cms-mcp": {
"url": "https://edge.vinkius.com/vk_preview_E2kDf3rmUJJgcNZKXaOKQyqaD4p2MZcV1xyGfKJa/mcp"
}
}
}List all the content models available in my Hygraph schema.
Show me the fields for the 'Product' model so I know what I can query.
Run a GraphQL query to get the title and slug for all entries in the 'Blog Post' model.
Find the ID for the entry named 'Summer Campaign' in my content library.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Hygraph once, then ask.
Just provide your Hygraph credentials. The connection is secure, and you can immediately start asking your AI to explore your content.
Connect Hygraph to your AIFAQ
How this task behaves
- 01
Can the AI change my content in Hygraph?
No. This connector is designed for reading and inspecting your schema and data. It cannot create, update, or delete content.
- 02
Does the AI write the GraphQL queries itself?
Yes. After it inspects your schema, it writes the queries needed to fetch the data you ask for.
- 03
Can the AI delete a model if I ask it to?
No. The tools provided only allow for schema introspection and data retrieval. It has no permission to modify your Hygraph structure.
- 04
How does the AI know which fields to use in a query?
It uses the schema introspection tools to see exactly which fields exist in your models before it attempts to run a query.
- 05
Can I see the raw GraphQL response?
Yes. Your AI client will show you the data it retrieved from the execution of the query.
More questions about Hygraph? See everything the Hygraph Connector can do
Connect Hygraph to Claude, Cursor, ChatGPT & more
