DatoCMS MCP. Control your entire content stack via conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
DatoCMS. This tool connects your AI client directly to your headless CMS. It lets you manage content records, execute GraphQL queries, and handle media assets—all via conversation.
You can list models, create new content entries, patch existing records, or delete old ones, all without leaving your chat interface.
It's full control over your digital content layer.
What your AI agents can do
Create cms record
Creates a new content item with a fully formed JSON payload.
Execute graphql cda
Finds specific data paths within the DatoCMS GraphQL tree for reading content.
Get media upload
Retrieves structured data that verifies specific file blocks.
Execute GraphQL queries to discover content paths and fetch data based on specific schemas.
Generate and provision new content items using a JSON Payload.
Safely change parameters on existing CMS records or delete them completely.
List, inspect, and retrieve details about uploaded media files and their storage identifiers.
List all registered content models and item types that define your content structure.
Retrieve explicit Cloud logging traces detailing CMS records using JSON:API.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
DatoCMS MCP Server: 10 Tools for Content & Media Ops
Use these tools to automate content creation, media retrieval, and record manipulation directly from your AI client.
019d7582create cms record
Creates a new content item with a fully formed JSON payload.
019d7582execute graphql cda
Finds specific data paths within the DatoCMS GraphQL tree for reading content.
019d7582get media upload
Retrieves structured data that verifies specific file blocks.
019d7582get single record
Extracts all properties and details for a single content node.
019d7582list cma records
Lists explicit records and logs, returning full JSON:API arrays.
019d7582list global models
Lists every structured rule that defines your content types and item definitions.
019d7582list media uploads
Inspects deep arrays of uploaded assets and media details.
019d7582patch cms record
Updates and modifies parameters on an existing content record safely.
019d7582wipe cms record
Permanently deletes a specific content record and its associated data.
019d7582wipe media upload
Removes a specific media asset after running a validation check.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with DatoCMS, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
This tool connects your AI client directly to DatoCMS. It lets you manage content records, run GraphQL queries, and handle media assets—all through conversation. You can list content models, create new content, patch old records, or delete them, all without leaving your chat interface. You get full control over your digital content layer.
- Querying Content Structure: Your agent runs GraphQL queries to find data paths and pull data based on specific schemas.
- Creating Content Records: You can generate and provision new content items using a JSON payload.
- Updating Content Records: You safely change parameters on existing CMS records or wipe them completely.
- Managing Media Assets: Your agent lists, inspects, and pulls details about uploaded media files and their storage identifiers.
- Auditing Content Models: It lists all registered content models and item types that define your content structure.
- Viewing Content Logs: You retrieve detailed logs about CMS records using JSON:API.
To use it, you subscribe to the server and plug in your DatoCMS Full Access API Token. Then, you just start managing your headless content from your AI client.
How DatoCMS MCP Works
- 1 Subscribe to the server and provide your DatoCMS Full Access API Token.
- 2 Instruct your AI client to perform a content operation (e.g., 'List all Blog Post records').
- 3 The server uses the token to communicate with DatoCMS, returning the requested data structure directly to your client.
The bottom line is you manage your entire content stack—from the database model to the final media asset—through one single chat interface.
Who Is DatoCMS MCP For?
Content Editors who can't leave the editor to update content or media. Front-end Developers who need to test GraphQL queries or fetch model definitions from the IDE. Product Managers who need to audit content versioning across environments. Digital Operations staff who must verify record consistency across the CMS.
Creating, updating, or managing media assets without switching tabs or leaving the content workspace.
Testing GraphQL queries and fetching content model definitions directly from the chat or IDE.
Monitoring content versioning and auditing media libraries across different environments for compliance.
Managing CMS models and verifying record consistency using natural language prompts.
What Changes When You Connect
- You can test complex data paths with
execute_graphql_cdadirectly in the chat, eliminating the need to jump back and forth between your IDE and the CMS UI. - Never lose track of assets again. Use
list_media_uploadsto inspect deep arrays and verify physical storage identifiers for every piece of media. - Need to update a page? Instead of clicking through 10 screens, use
patch_cms_recordto modify specific attributes on a record and commit the changes immediately. - Starting fresh? Use
create_cms_recordto generate a complete, valid content item payload from scratch, hitting the database with one command. - Need to know what content exists? Run
list_global_modelsto get an enumeration of all registered content types. It shows you the structure before you even write the query. - When a record is dead, use
wipe_cms_record. This permanently vaporizes the node, clearing out the data when you need it gone.
Real-World Use Cases
The site needs a new blog post.
A content editor needs to publish a post, but the model requires a new 'Author' record and multiple 'Product' references. Instead of manually creating records in two different sections, the agent runs create_cms_record to provision the entire JSON payload, ensuring all required fields are present before publication.
I need to check if the image is optimized.
A digital ops engineer needs to confirm the storage ID and proxy mapping for a hero image. They run get_media_upload and receive the exact structural match, confirming Imgix is set up correctly before development starts.
A client changed the required content model.
A front-end developer needs to know exactly what fields are available for the 'Product' model. They run list_global_models to get a complete list of available types and their definitions, saving hours of manual exploration.
A record is corrupted and must be removed.
A product team member finds an old, test document that needs to be completely erased. They use wipe_cms_record to irreversibly vaporize the document node, guaranteeing no stale data remains in the database.
The Tradeoffs
Chaining multiple read calls
Getting a single record requires calling get_single_record first, then manually running execute_graphql_cda to find the related image path, and finally calling get_media_upload to see the asset details. This is slow and requires 3-4 separate tool calls.
→
The agent handles this sequence. You ask for 'the details of the main product page, including the primary hero image.' The agent combines the data flow, using get_single_record and get_media_upload in one transaction to deliver the full payload.
Manually cleaning up records
When an old content type is deprecated, a developer might try to manually delete records by calling list_cma_records to get IDs, and then running wipe_cms_record for each one. This is tedious and prone to missing nodes.
→ The agent can target the entire scope. You prompt: 'Delete all records for the deprecated 'Test' model.' The agent executes the cleanup against the model scope, handling all necessary deletions.
Using the wrong data mode
Attempting to modify content structure (CMA) while only querying for display data (CDA). This results in incomplete patches or failed writes because the correct context isn't set.
→ The tool handles the switch. You tell the agent to 'Update the schema for Blog Post records.' The agent automatically switches to the Content Management (CMA) mode to ensure structural edits succeed.
When It Fits, When It Doesn't
Use this server if your workflow requires managing content structure (models, types) and media assets and you need to perform full CRUD operations without leaving your AI chat client. Point is: you need to treat the CMS like a programmatic service layer. Don't use this if you only need to read a simple, pre-defined list of posts—a simple GraphQL query wrapper might suffice. If your primary need is just simple logging or analytics, use a dedicated logging tool instead. When you need to modify data and check the schema, use this. For example, if you need to patch_cms_record, this is the right place. If you only need to execute_graphql_cda for reading, you still use this, because the ability to write is what makes it valuable.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by DatoCMS. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Sifting through content models and data paths is a massive time sink.
Right now, if you want to know what fields are available for a 'Product' page, you have to navigate the DatoCMS dashboard, click into the model definition, and scroll through dozens of fields. If you want to change a field type, you have to switch to a dedicated content management view, and if you want to run a query, you jump to the GraphQL Playground. It’s a three-app workflow, and you're copying data between all three.
With the DatoCMS MCP Server, you just tell your agent: 'List all content models and tell me what fields the Product type has.' The agent runs `list_global_models` and gives you the structure right there. You get the schema definition and the full history of the record, all from the chat window.
DatoCMS MCP Server: Full control over content and media.
Instead of performing a read operation, then manually finding the ID, then running a separate update query, you just ask the agent to 'Update the product price to $199 and attach the new banner image.' The agent handles the `patch_cms_record` call and the `get_media_upload` reference internally.
The difference is this: you don't manage the workflow; the agent does. You state the desired end state, and the server handles the complex, multi-step transactional logic to get there. It's one command.
Common Questions About DatoCMS MCP
How do I list content models using the DatoCMS MCP Server? +
Use the list_global_models tool. This tool gives you an enumeration of every structured rule and item type defined in your content, letting you see exactly what data structures you're working with.
Can I delete a record using the DatoCMS MCP Server? +
Yes, you use wipe_cms_record. This tool permanently vaporizes a document node, ensuring the record is removed from your live data set.
How do I run a GraphQL query with the DatoCMS MCP Server? +
Use execute_graphql_cda. This tool identifies bounded routing spaces inside the GraphQL tree and runs the query against your content delivery data.
What does `patch_cms_record` do? +
patch_cms_record safely updates existing records. It lets you modify specific item parameters without overwriting the entire document payload.
How do I find media uploads with the DatoCMS MCP Server? +
Use list_media_uploads. This tool inspects deep internal arrays, giving you a list of all uploaded assets and their metadata.
How do I authenticate and set up the DatoCMS MCP Server? +
You authenticate by providing your DatoCMS Full Access API Token, which you find in Settings > API Tokens. Your AI client uses this token to manage content, ensuring full, secure access to your headless CMS.
What is the difference between `list_cma_records` and `execute_graphql_cda`? +
The execute_graphql_cda tool runs read-only queries against the Content Delivery API, identifying routing spaces. In contrast, list_cma_records retrieves explicit JSON:API arrays directly from the Content Management API for structural edits.
How can I manage content versioning using the DatoCMS MCP Server? +
The server allows monitoring content versioning and auditing media libraries across different environments. You can manage these details through natural language prompts, keeping your content process consistent.
Can my agent execute custom GraphQL queries against DatoCMS? +
Yes. Use the 'execute_graphql_cda' tool. You can provide any valid GraphQL query string to fetch data from the Content Delivery API (CDA) tree, bypassing rigid limits and retrieving exactly what your agent needs.
How do I create a new content item using the agent? +
Use the 'create_cms_record' tool. Provide the Item Type ID and a JSON object containing the attributes. The agent will orchestrate the absolute explicit bindings to generate a new record in your DatoCMS project.
Can I inspect the available content models through chat? +
Absolutely. The 'list_global_models' tool enumerates all registered item types and models. This allows your agent to identify precisely what schemas bind to your editor blocks and which fields are available for data mutations.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Lorem Picsum
Generate placeholder images and retrieve image metadata directly from your AI agent using the Lorem Picsum API.
Coqui TTS (Open Source Speech Studio API)
Generate high-quality speech from text using Coqui TTS. List available models and synthesize audio directly from your AI agent.
Vercel Alternative
Manage Vercel deployments, projects and domains via API — inspect builds, track deployments, manage env vars and domains from any AI agent.
You might also like
Razorpay
Automate online payments via Razorpay — manage orders, payments, and refunds directly from any AI agent.
Parknav
AI-powered predictive parking availability and street occupancy data via Parknav API.
Scale SERP
Equip your AI agent with real-time Google Search data — run organic, news, shopping, and scholar queries effortlessly.