Payload CMS MCP. Control your content and database logic via AI.
Payload CMS MCP lets your AI client interact with complex headless content management systems directly. It provides structured tools for reading, writing, and updating data across collections, singletons, and custom schemas without needing hand-written API endpoints.
Claude 
ChatGPT 
Cursor 
Gemini 
Windsurf 
VS Code 
JetBrains 
Vercel Give Claude and any AI agent real-world access
The MCP writes brand new JSON documents into specified Payload collections.
It patches specific fields within a document or collection using its unique ID.
You can extract configuration details from global, singular items that control the entire website.
The MCP scans and retrieves the list of structured data types available in your CMS.
It queries existing user records to check permissions or find active admin accounts.
Ask an AI about this
Waiting for input…
What AI agents can do with Payload CMS MCP: 10 Tools Available
These tools let your agent perform all standard database operations—from creating new records to checking global configurations—directly within the Payload CMS structure.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Payload CMS MCPCreate Cms Document
Write new, fully structured JSON documents into designated Payload collections.
Wipe Cms Document
Permanently remove specific document rows from the active CMS collection.
Get Single Document
Inspect deep internal arrays to retrieve data for a single, specified record.
Get Singleton Global
Extract properties that control global settings or site-wide configurations.
List Collection Documents
Find and list all the available structured data types within your CMS collections.
List Payload Users
List active user accounts, checking which admin identities are currently set up.
Verify Token Identity
Check the current user's identity and permissions associated with the API token.
Search Collection Where
Perform advanced searches across a collection based on specific criteria or filters.
Patch Cms Document
Modify and update existing documents by substituting specific database blocks using...
Update Singleton Global
Dispatch an automated check to ensure global settings are correctly updated across...
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Payload CMS integration is available immediately — no restart needed.
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 each call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Payload CMS, then connect any of our 5,200+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,200+ others, all in one place
- Add new capabilities to your AI anytime you want
- Connections are secured and governed automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog weekly
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Payload CMS. 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 CLOUD
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on each call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Managing Content in Headless CMSs Is Messy Work
Right now, if your team needs to audit the site's current configuration, you have to jump into three different areas: checking global singletons for primary settings, listing collections to see what content types exist, and then running separate queries just to pull a few specific documents. It’s tedious clicking through dashboards and copy-pasting IDs.
With this MCP, your agent handles the complexity of those database calls automatically. You simply ask it to check the site's global settings or list all available document types. The result is clean data that you can use immediately in your next step.
Payload CMS: Control Content with Write Operations
The biggest time sink used to be the manual process of updating a document. You'd find an ID, copy it, open the correct collection view, patch the JSON payload, and hope you didn't break something else in the process.
Now, your agent manages that whole cycle. You tell it to update the content using `patch_cms_document`, provide the target ID, and define the new data. The job is done; no manual clicking required.
What Payload CMS MCP does for your AI
You can tell your AI agent to manage your website's backend data as if it were a native interface. Instead of building complex code just to fetch or write content, you simply ask for the action—like listing all posts in a collection or updating a specific JSON field on a profile document.
Your agent handles the Payload CMS logic automatically, letting you focus on generating the final output, not managing the database plumbing. For example, if you need to update a global site setting, your AI client can execute that change with minimal prompts. Connecting this MCP via Vinkius gives your agent access to over 4,000 other tools, meaning it's one place for all your data needs.
You write the prompt; your AI handles the database interaction.
019d75f0-1783-72bc-bcd9-507139d9a2aa 
How to set up Payload CMS MCP
The bottom line is you get a natural language way to control and manipulate complex database structures.
First, you subscribe and provide your specific Payload Base URL and private API token parameters.
Next, you tell your AI client exactly what data you need—for example, 'list all documents in the posts collection' or 'update the site title to X'.
The MCP runs the command against your CMS, returning structured JSON data that your agent can immediately use for the next step.
Who uses Payload CMS MCP
This MCP is crucial for backend developers, content engineers, and frontend teams who need to test or modify data structures without constantly dropping into the CMS admin interface. If your job involves reading, updating, or validating site content via API calls, this is what you need.
You use it to manipulate complex database items and run deep schema revisions when testing new features.
You map dynamic content updates directly, applying JSON payload changes across multiple collections for site staging or pre-launch checks.
You isolate singleton configuration outputs to validate the global state of your API during response testing.
Benefits of connecting Payload CMS MCP
Instead of writing boilerplate code to read data, you can simply ask the agent to list all documents in a collection using list_collection_documents. This saves hours of repetitive API setup.
Need to make a change? Use patch_cms_document to modify fields on an existing document without having to write full update scripts. It targets specific blocks by ID, making the action precise.
The MCP handles complex validation for you. If your site relies on global settings, use get_singleton_global to pull that configuration state and test your logic against it immediately.
Content lifecycle management gets easier with tools like create_cms_document. You can instruct the agent to write a new post or category record simply by providing the JSON data payload.
Don't waste time figuring out who has access. Use list_payload_users and verify_token_identity to audit your system's user roles before deploying any major changes.
Payload CMS MCP use cases
Staging a new feature page
The content engineer needs to test how the site handles 20 different product descriptions. Instead of manually running 20 API calls, they ask their agent to run list_collection_documents for 'products' and then loop through the results, using get_single_document on each one to validate its structure.
Fixing a broken site setting
The frontend team notices the primary navigation header is wrong. They ask their agent to use get_singleton_global to retrieve the current global settings, spot the error, and then use patch_cms_document to correct just the title field.
Deprecating old content
An admin needs to remove an entire outdated media gallery. Instead of running a complex deletion script, they tell their agent to target and wipe the specific record using wipe_cms_document by its ID.
Auditing user permissions
The security team wants to know if only admins can create new content. They prompt their agent to run list_payload_users, then use the results to check which roles have permission to perform a write operation.
Payload CMS MCP tradeoffs
What to watch out for, and the recommended way to handle each one.
Treating the CMS like simple key/value store
Attempting to update global site settings by just providing two fields, assuming the API will handle everything. This often fails because singletons have complex relational structures.
Always use get_singleton_global first to understand the full schema before making changes. Then, use patch_cms_document with all required IDs and structured data to ensure a complete update.
Manually writing every CRUD endpoint
Writing separate Python functions for listing, creating, and patching documents just because the CMS is complex.
Use this MCP. The agent handles the complexity of the Payload REST API calls when you simply ask to list_collection_documents or create_cms_document.
Ignoring content schemas
Trying to write a document using data that doesn't match the required fields for the collection, causing the whole operation to fail.
Use list_collection_documents first. This confirms what structure you're dealing with and helps your agent generate correctly formatted JSON payloads.
When to use Payload CMS MCP
You should use this MCP if your workflow requires interacting with the complex, structured data of a headless CMS—specifically needing to read global settings, create new documents, or patch existing records based on their ID. It's perfect for testing and migration logic.
Don't use it if you just need simple data fetching (e.g., 'get all user names'). If your needs are purely additive or require complex external business logic that doesn't touch the CMS, a general API connector might be better. But when the core problem is 'How do I tell my AI agent to write content to this specific Payload structure?' then you need this MCP.
Frequently asked questions about Payload CMS MCP
How do I find out what collections exist in Payload CMS using the Payload CMS MCP? +
You use list_collection_documents. This tool scans your entire setup and returns a list of all structured data types, letting you know exactly where to look for content.
Can I update global site settings with the Payload CMS MCP? +
Yes. You use patch_cms_document or update_singleton_global. These tools are designed specifically to manage those crucial, site-wide configuration items.
Is the payload cms mcp safe for deleting content? +
It is safe when used correctly. The MCP provides wipe_cms_document to irreversibly delete specific records after your agent has verified which documents need removal.
Does Payload CMS MCP work with my private API key? +
Yes, you provide the required Private API Token when setting up the connection. This ensures that only authorized calls can manipulate your backend data.