TiDB Cloud MCP. Inventory projects, clusters, and instances instantly.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
TiDB Cloud (Serverless Distributed SQL) MCP Server lets you manage your entire database infrastructure through chat. You can list every project, check specific cluster details, and pull a full inventory of all TiDB X instances across different tiers—all without opening the web console.
What your AI agents can do
Get project
Retrieves detailed information for a single specified project.
List clusters
Gets a list of all dedicated database clusters within a specific project.
List instances
Pulls an inventory of all TiDB X instances running inside your projects.
Retrieves a full list of every accessible project within your TiDB Cloud organization.
Fetches detailed metadata and IDs for one specified project using its name or ID.
Retrieves a complete inventory of every TiDB X instance (Starter, Essential, Premium) housed across your projects.
Lists and provides details on dedicated cluster topologies for high-performance database inspection.
Ask AI about this MCP
Supported MCP Clients
Waiting for input…
TiDB Cloud (Serverless Distributed SQL) MCP Server: 4 Tools for DB Ops
These four tools let your AI client manage resource scoping by listing projects, getting details, cataloging instances, and inspecting dedicated clusters.
019e5d60get project
Retrieves detailed information for a single specified project.
019e5d60list clusters
Gets a list of all dedicated database clusters within a specific project.
019e5d60list instances
Pulls an inventory of all TiDB X instances running inside your projects.
019e5d60list projects
Returns a list of every project accessible in your entire organization.
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 TiDB Cloud (Serverless Distributed SQL), 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
You don't gotta open up a console or click through ten different dashboards to manage your TiDB Cloud setup anymore. Just connect your AI client, and your agent handles all the dirty work of talking directly to your database infrastructure. This server gives you command-line access to everything running in your organization.
If you need an overview, start by getting a full map of where your resources live. You can run list_projects to see every single project accessible across your entire TiDB Cloud account. That's your master list. Once you know which environment you’re dealing with, you use get_project to pull the detailed metadata and specific IDs for any one project by its name or ID.
This gives you a deep dive into that environment’s setup without leaving the chat window.
When it comes to your actual data stores, checking inventory is simple but crucial. Running list_instances pulls up a complete count of every TiDB X instance—you're talking Starter, Essential, and Premium tiers—that you've got running inside all those projects. This single call gives you the full scope of what’s active right now.
You won't miss a thing.
For performance-critical stuff, you need to inspect the cluster topology. You use list_clusters to get a list and details on dedicated database clusters within a specific project. These are your high-performance setups; this tool lets you check their status and structure quickly for auditing or planning upgrades. It's key for understanding how your core data is actually laid out.
The whole point here is that you bypass the UI friction. Instead of clicking through menus, your agent uses these tools to extract precise information: it gets the project names needed for scripting, pulls specific resource counts from list_instances, or maps out the physical structure using list_clusters. It streamlines checking whether a particular instance needs patching or if a whole project's metadata is up-to-date.
You simply tell your agent what you need to know—like "Show me all Premium instances in Project Alpha"—and it runs the necessary sequence of calls and spits out the actionable data right here. You’ll save yourself hours of clicking around in the web console just managing IDs, checking resource allocation across multiple environments, or confirming that every single required cluster is online and healthy.
How TiDB Cloud MCP Works
- 1 Subscribe to the server and input your TiDB Cloud Public and Private API Keys.
- 2 Your AI client (Claude, Cursor, etc.) runs a command like 'list all projects.'
- 3 The MCP Server calls the appropriate tool (
list_projects) and returns structured data that your agent uses for conversation.
The bottom line is: you send an intent to your agent, and it executes the correct API call using these tools.
Who Is TiDB Cloud MCP For?
Anyone who spends time clicking through web consoles just to answer a simple inventory question. This is for the DBA who needs to audit 50 clusters across 12 projects, or the backend engineer who needs to verify a specific Project ID before writing migration code. It's built for people whose job involves finding data buried deep in cloud dashboards.
Audits cluster distributions and project structures, checking which projects house Starter vs. Premium instances without logging into the web console.
Verifies instance availability, checks project IDs, or confirms resource existence directly from their coding environment during development cycles.
Automates the discovery of cloud resources for status reporting, documentation generation, or pre-deployment validation scripts.
What Changes When You Connect
- Audit resource scope fast. Instead of clicking through menus to see what's running where, use
list_projectsfirst. You get a flat list of all organizational containers so you know exactly where your data lives. - Check every instance in one go. The
list_instancestool pulls the full inventory—Starter, Essential, Premium—so you don't have to guess which cluster has which service tier. - Pinpoint specific resource details. Need a Project ID for a script? Run
get_projectto pull deep metadata without manual copy-pasting from a dashboard view. - Audit high-performance setups. Use
list_clustersto get a clear, immediate breakdown of dedicated cluster topologies. Perfect for pre-migration checks. - Eliminate web console hopping. You keep all your resource data—project IDs, instance counts, cluster names—inside your chat window, making debugging much faster.
Real-World Use Cases
Debugging a deployment failure
The new backend engineer runs into an API error related to a missing resource ID. Instead of asking a colleague, they prompt their agent: 'What are all the project IDs?' The agent uses list_projects, providing them with the exact list needed to resolve the bug immediately.
Quarterly infrastructure audit
A DBA needs to report on resource utilization for compliance. They ask their agent to 'List all projects and count instances.' The agent chains list_projects then calls list_instances per project, giving the DBA a comprehensive, structured report.
Pre-migration validation
A DevOps team is planning a major database upgrade. They run 'Show me all dedicated clusters' using list_clusters. This confirms that every critical high-performance topology exists and is accounted for before they start the migration scripts.
Resource scoping check
An engineer wonders if a specific microservice runs in 'Staging'. They ask, 'What projects exist?' The agent uses list_projects, allowing the engineer to visually confirm that the 'Staging-Env' project exists and then drill down with get_project.
The Tradeoffs
Manual web console clicking
The DBA opens the TiDB Cloud dashboard, clicks 'Projects,' copies a list of IDs. Then they navigate to 'Instances' and copy those results into a spreadsheet to cross-reference.
→
Just ask your agent: 'List all projects, then pull all instances.' The tool chain handles the linking using list_projects followed by list_instances, giving you structured data in one go.
Assuming scope
An engineer assumes a resource is in Project X because they saw it months ago. They spend time debugging why the API call fails, not realizing they need to confirm the ID first.
→
Always validate with get_project before relying on an ID or project name. This confirms the resource scope and ensures your subsequent calls are pointed at the right place.
Writing custom scripts for simple lists
A developer writes a 50-line Python script just to get 'all accessible projects' because they think an API call is too simple.
→
Use the list_projects tool. It handles the basic listing logic instantly, letting you focus on complex data processing instead of boilerplate connection code.
When It Fits, When It Doesn't
Use this MCP Server if your core task involves auditing, inventorying, or retrieving metadata about multiple interconnected database resources (Projects -> Clusters -> Instances). You need to answer questions like: 'Which projects have Premium instances?' or 'List every cluster in the Production group.'
Don't use it if you only need to run a single query against one specific table. If your task is just running SELECT * FROM users WHERE id = 5, stick with direct database connectivity tools. This tool is for finding and organizing resources, not querying their contents.
If your goal requires cross-resource queries (e.g., 'find all instances exceeding 80% CPU usage'), this set of four tools gives you the foundation, but you'll need client-side logic to chain them together.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by TiDB Cloud. 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 4 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Trying to inventory a cloud database feels like clicking through three different tabs.
Today, checking your environment is tedious. You open the dashboard, click 'Projects' just to get a list of IDs. Then you have to switch tabs for 'Instances,' and then maybe into another section for 'Dedicated Clusters.' It’s a multi-step process that forces context switching and copy-pasting.
With this MCP Server, those steps vanish. You tell your agent what you need—say, 'List all projects and their instances.' The tool handles the full workflow behind the scenes, giving you structured data right in your chat window. It's instant resource mapping.
TiDB Cloud (Serverless Distributed SQL) MCP Server: Get project metadata instantly.
Manual auditing requires jumping between the Project view, the Instance list, and the Cluster details. You waste time verifying if a resource exists before you can even start your analysis.
Now, use `get_project` or `list_projects`. Your agent pulls that metadata directly to your client. The data is where it needs to be: right here.
Common Questions About TiDB Cloud MCP
How do I list all the projects using the get_project tool? +
You don't use get_project to list everything; that tool requires a specific ID. To see all available scopes, you must run list_projects. That function gives you the full organizational overview first.
Can I find out every single running instance with list_instances? +
Yes, list_instances pulls a complete inventory of all TiDB X instances—regardless of whether they're Starter or Premium. It gives you the full picture in one call.
What is the difference between list_clusters and list_instances? +
They track different things. list_instances handles individual database nodes (the X instances). list_clusters focuses specifically on grouping those resources into a high-performance Dedicated cluster topology.
Does this MCP Server help me check project details? +
Absolutely. Use the get_project tool and provide the specific Project ID you need. It retrieves all relevant metadata for that single, specified environment.
What credentials do I need to use the list_projects tool? +
You must provide both your TiDB Cloud Public and Private API Keys when setting up the server. These keys allow your AI agent to authenticate against your account, ensuring you only see resources you have access to.
If I run list_clusters for a project, but no dedicated clusters exist, what response should I expect? +
The tool returns a specific, non-error message stating that zero clusters were found. This clear output allows your agent to proceed with other tasks without assuming failure.
When using list_instances, can I filter the results by database tier (e.g., Premium)? +
Yes, you pass a scope parameter to refine instance retrieval. This allows you to pull data only for specific tiers—like Starter or Essential—without processing unrelated records.
Are there any rate limits when calling list_projects? +
The server adheres strictly to TiDB Cloud’s API rate limits. If you exceed the allowed number of calls, your agent will receive a standard 429 error code until your usage window resets.
Can I see both serverless and dedicated clusters with this server? +
Yes. Use list_instances to see TiDB X instances (Serverless/Starter/Premium) and list_clusters to view your Dedicated cluster infrastructure.
How do I get the ID of a specific project to query its instances? +
You can run the list_projects tool first. It will return a list of all accessible projects along with their unique IDs, which you can then use in list_instances or list_clusters.
Does this tool allow me to delete or create new clusters? +
No. The current version of the TiDB Cloud MCP server is focused on discovery and inspection (read-only operations) such as list_projects, get_project, and list_clusters to ensure safe infrastructure monitoring.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Nhost
Manage Nhost authentication and storage — handle user sign-ins, registrations, profile management, and file uploads directly from any AI agent.
Baserow
Build no-code databases, create custom views, and collaborate on structured data with an open-source Airtable alternative.
Snowtrace (Avalanche Explorer)
Explore the Avalanche C-Chain — check AVAX balances, track ERC-20/NFT transfers, and inspect smart contract data directly from your AI agent.
You might also like
Sender.net
Send email and SMS campaigns that convert with an affordable marketing platform built for e-commerce and small businesses.
NRC ADAMS (Nuclear Regulatory)
Access official Nuclear Regulatory Commission (NRC) documents, news, and reactor status reports directly through the ADAMS database.
Unpaywall (Open Access Papers)
Access millions of open-access scholarly articles. Check OA status and retrieve full-text PDF links using DOIs.