Gridscale MCP. Manage cloud servers, IPs, and networks via chat.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Gridscale (IaaS & PaaS Cloud Hosting API) MCP Server lets you manage your cloud infrastructure directly from your AI agent.
You can list servers, create new VMs, check CPU metrics, and control network links—all via natural language. Use it to provision test environments, monitor resource health, or manage firewalls without logging into a console.
What your AI agents can do
Create server
Builds and provisions a brand new cloud server instance.
Get server metrics
Retrieves real-time data, like CPU usage and storage IOPS, for a specific server.
Get server power
Checks the current on/off power status of a specified server.
Create new virtual machines using the create_server tool.
Get the power status of a server using get_server_power, or retrieve performance metrics with get_server_metrics.
List all IPs (list_ips), networks (list_networks), and firewalls (list_firewalls) to manage connectivity.
Link storage volumes to servers (link_storage_to_server) or attach an IP address (link_ip_to_server) to an existing machine.
Start, stop, or perform an ACPI shutdown on a server using set_server_power or shutdown_server.
Access comprehensive lists of PaaS services (list_paas), storage volumes (list_storages), or available locations (list_locations).
Ask AI about this MCP
Supported MCP Clients
Waiting for input…
Gridscale MCP Server: 17 Tools for Cloud Control
Use these tools to manage everything from server creation and power status checks to linking storage volumes and managing network firewalls.
019e5d22create server
Builds and provisions a brand new cloud server instance.
019e5d22get server metrics
Retrieves real-time data, like CPU usage and storage IOPS, for a specific server.
019e5d22get server power
Checks the current on/off power status of a specified server.
019e5d22link ip to server
Connects a specific IP address to a server.
019e5d22link storage to server
Attaches a storage volume to a server; the server must be powered off first.
019e5d22list deleted
Shows a list of objects that have been deleted, useful for tracking history.
019e5d22list firewalls
Lists all existing firewall rules and configurations.
019e5d22list ips
Gets a full list of all IP addresses assigned in your account.
019e5d22list iso images
Shows all available ISO images for server deployment.
019e5d22list locations
Lists available data center locations (e.g., Frankfurt) where you can deploy resources.
019e5d22list networks
Gets a list of all private networks in your account.
019e5d22list paas
Retrieves a list of all managed platform services (PaaS) like databases.
019e5d22list servers
Displays a list of every server currently in your Gridscale account.
019e5d22list storages
Displays a list of all available storage volumes.
019e5d22list templates
Shows all available deployment templates.
019e5d22set server power
Changes the power state of a server (e.g., turn it on or off).
019e5d22shutdown server
Performs a complete ACPI shutdown on a server.
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 Gridscale (IaaS & PaaS Cloud Hosting API), 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
Gridscale API MCP Server - Manage Cloud Hosting lets your AI agent handle your whole cloud setup. You don't gotta log into a console to manage your infrastructure; you just talk to it. You can provision new servers, check resource health, and tweak networking rules using natural language. You'll find tools for managing both IaaS virtual machines and PaaS services.
Provisioning and Status Checks
You can build a brand new virtual machine using create_server. To check on an existing machine, use get_server_power to see if it's on or off, or get_server_metrics to pull real-time data like CPU usage and storage IOPS. You can also shut down a server completely with shutdown_server or change its power state using set_server_power. Managing Resources
To handle networking, your agent can list all assigned IP addresses with list_ips, all private networks with list_networks, and every existing firewall rule using list_firewalls. You can connect resources to machines; use link_storage_to_server to attach a storage volume, remembering you gotta power the server down first. You connect an IP address using link_ip_to_server to an existing machine. Getting the Picture
Need to know what else is running? You can pull lists of all available PaaS services with list_paas, or check out all available storage volumes with list_storages, and see all the deployment templates with list_templates. You can also see which data centers you can use with list_locations, or what ISO images are available for deployment using list_iso_images. Server Inventory and History
Your agent can display a list of every server currently in your account using list_servers. You can list all available network configurations by checking list_firewalls and all network types by running list_networks. You can also get a list of objects that were deleted using list_deleted. Core Operations
To manage your server state, you can run set_server_power to toggle the power, or shutdown_server to perform an ACPI shutdown. You can build a new machine with create_server. You can attach storage to a server using link_storage_to_server after you've powered it down. You connect an IP address to a server with link_ip_to_server. Monitoring and Scope
To check a server's health, run get_server_metrics for CPU usage and storage IOPS. To see all the things you're working with, you can check the list of available storage volumes with list_storages or the list of available PaaS services with list_paas.
How Gridscale MCP Works
- 1 Subscribe to the server and provide your Gridscale User ID and API Token.
- 2 Your AI client sends a natural language command (e.g., 'Check the CPU usage for Production-Web').
- 3 The MCP server invokes the correct tool (e.g.,
get_server_metrics) and returns the structured data to your AI agent for action.
The bottom line is, you talk to your AI agent, and it executes the infrastructure commands for you.
Who Is Gridscale MCP For?
Ops engineers who hate clicking through dashboards at 2am. Developers building test environments who need to spin up resources fast. System admins who manage networking and resource compliance. If you spend time logging into a cloud console to check status, this is for you.
Automates infrastructure audits, checks resource metrics, and restarts failing services without leaving the terminal or chat window.
Quickly provisions isolated test environments and verifies the status of managed databases during the coding process.
Manages complex network configurations, updates firewall rules, and monitors resource utilization through simple conversation.
What Changes When You Connect
- Stop logging into the console. You manage the full server lifecycle—from
create_servertoshutdown_server—all through your AI agent's natural conversation. - See real-time performance data instantly. Use
get_server_metricsto check CPU usage and storage IOPS without running a separate dashboard query. - Automate network setup. Use
list_ipsandlist_firewallsto check available IP ranges and validate security rules before linking them to a new server. - Handle resource dependencies easily. You can run
link_storage_to_serverorlink_ip_to_serverby just asking, preventing the manual steps of finding resource IDs. - Get a full picture of your stack. Run
list_paasto see managed services, andlist_serversto get a comprehensive inventory list in one go. - Avoid manual provisioning delays. Use
list_locationsto pick the right data center, thencreate_serverto spin up resources faster.
Real-World Use Cases
Initial Audit: Checking the entire fleet's health
An ops engineer needs to know if 15 servers are healthy and if any IPs are running low. They ask their agent to run list_servers and then follow up with get_server_metrics for the top 5 candidates. The agent compiles the inventory list and the performance data, flagging any server whose CPU usage is above 80%.
Debugging a Connectivity Failure
A developer reports that the staging environment can't talk to the database. They ask the agent to run list_firewalls and list_networks. The agent identifies a restrictive rule blocking port 5432, allowing the developer to modify the rule using list_firewalls and link_ip_to_server to correct the connection.
Rapid Test Environment Setup
A developer needs a fresh VM for testing a new microservice. They ask the agent to first use list_templates to pick a base image, then create_server with the chosen template, and finally use link_storage_to_server to attach a dedicated volume. The agent executes the sequence, minimizing setup time.
Scheduled Maintenance Shutdown
A system admin needs to shut down 20 servers for maintenance. They tell the agent to first run list_servers to confirm the target list, then loop through the list and call shutdown_server for each one, confirming the power state with get_server_power afterward.
The Tradeoffs
Trying to list everything in one shot
Running a single command to 'list all resources' leads to a massive payload, often hitting API limits or causing the connection to time out, leaving you with partial data.
→
Always break down resource discovery. Use specific calls like list_ips to check IPs, and then list_firewalls to check the rules. If you need to see everything, run list_servers and then individually call list_paas for the managed services.
Assuming resources are available
A user tries to link_ip_to_server without first checking if the IP is free, resulting in an immediate failure and wasting time troubleshooting a basic dependency issue.
→
Before linking, validate the resource. Always run list_ips to confirm the IP is available, and if necessary, use list_storages to ensure the required volume exists.
Modifying state without knowing the current state
Attempting to link_storage_to_server when the server is still powered on, which will result in a validation error and require a manual power-down step.
→
Always check the power status first. Use get_server_power or list_servers to verify the server is powered off before attempting to link storage.
When It Fits, When It Doesn't
Use this MCP Server if your workflow involves managing interconnected, physical cloud resources (VMs, IPs, Storage, Firewalls). It's perfect for the ops engineer who needs to run a multi-step audit: check list_servers > check get_server_metrics > validate security via list_firewalls > act via link_ip_to_server. Don't use it if you only need to read general data (like a list of users). If your problem is simply 'I need a list of things,' use a dedicated listing tool. If your problem is 'I need to know if this resource is healthy,' use get_server_metrics first. Never assume the process is single-step; it's always a chain of calls.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Gridscale. 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 17 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Manually checking server status and resource links is a time sink.
Right now, you open the console. You click 'Servers.' You find the ID. You check the status tab. Then you open the Networking tab. You cross-reference the IP range. If you need to attach storage, you have to switch to the Storage section, find the volume ID, and then manually link it. It's a dozen clicks and three tabs.
With this MCP server, you tell your agent: 'Check the metrics for the web server and ensure it has the correct IP attached.' The agent runs `get_server_metrics`, validates the IP using `list_ips`, and confirms the link—all in one conversation. You get the answer, not a stack of links to click.
Gridscale MCP Server: Control infrastructure state.
You don't have to SSH into a machine just to check if it's running or if you need to shut it down. You don't have to remember the specific API endpoint for power control. You just ask your AI client to set the power state using `set_server_power` or to perform a hard shutdown with `shutdown_server`.
The difference is simple: you speak natural language, and the infrastructure changes. No more navigating complex menus or remembering arcane CLI commands. It just works.
Common Questions About Gridscale MCP
How do I use the `get_server_metrics` tool to monitor performance? +
You specify the server ID and the time window. The agent fetches the real-time metrics, providing CPU usage and storage IOPS data. This is better than checking a dashboard because you get the data immediately in your chat stream.
Can I use `link_storage_to_server` without powering off the server? +
No, the tool requires the server to be powered off. You must first use get_server_power to confirm the status, and if needed, use set_server_power before running link_storage_to_server.
Which tool do I use to see all my networking rules? +
Use list_firewalls. This tool lists all firewall rules. You can then use list_networks to see the associated network IDs for context.
How do I create a new server using `create_server`? +
You provide the required specifications, such as the location and resource size. The agent handles the required resource linking steps, which is much easier than running a multi-parameter CLI command.
What is the difference between `shutdown_server` and `set_server_power`? +
set_server_power changes the power state (on/off). shutdown_server forces an ACPI shutdown, which is a more complete, controlled termination process.
How do I list all available data centers using `list_locations`? +
Use list_locations to get a list of available datacenters/locations. This is crucial when creating new resources, as you need a valid location UUID to proceed.
What tools handle networking resources like IP addresses and firewalls? +
You use list_ips to see all IP addresses and list_firewalls to review existing firewall rules. These tools help you manage network configuration details.
What is the purpose of `list_deleted`? +
list_deleted tracks historical data by showing deleted objects. This is useful for auditing and retrieving records you no longer have immediate access to.
Can I check the power status of a specific server using its UUID? +
Yes! Use the get_server_power tool with the server's UUID. The agent will return whether the server is currently powered on or off.
How do I monitor the performance of my cloud servers? +
You can use the get_server_metrics tool. It retrieves data in 15-minute intervals for core usage and storage IOPS, allowing your AI to analyze resource consumption.
Does this integration support managed databases like MySQL or PostgreSQL? +
Yes, you can list all managed Platform Services (PaaS) including MySQL, PostgreSQL, and Redis using the list_paas tool.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
GeekJoke
Get random geeky and programming-related jokes to lighten up your development workflow.
Context7
Empower AI agents via Context7 — pull up-to-date documentation and code examples for any library or framework directly into your workspace.
Markdown Frontmatter Harvester
Scan local Obsidian or Hugo vaults and extract all YAML frontmatter (tags, dates, status) into a single structured JSON. Let your AI query your scattered notes metadata instantly.
You might also like
Buildium
Manage rental properties via Buildium — track units, tenants, and maintenance tasks directly from any AI agent.
Namecheap
Manage your domains, DNS records, SSL certificates and account — check domain availability and configure nameservers.
Guance Cloud / 观测云
Modern observability platform — manage monitors, dashboards, and events via AI.