Civo MCP for AI. Manage K8s clusters and compute resources via chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Civo (Cloud-native Kubernetes Cloud Provider API) connects your agent directly to Civo's infrastructure APIs. Provision K3s clusters, manage compute instances, and configure networking rules—all through natural language chat.
Check resource quotas, attach persistent volumes, and monitor usage without leaving your IDE.
What your AI can do
Add team member
Adds a specified user account to an existing team structure within Civo.
Attach volume
Connects a pre-created block storage volume to a running compute instance for persistent data.
Create cluster
Initiates the process of creating an entirely new Kubernetes cluster within your account.
Create new K3s clusters in different regions or list existing ones for auditing.
Launch, stop, reboot, resize, or tag virtual machines to fit your workload needs.
Establish private networks and apply granular firewall rules to restrict traffic between resources.
Create block storage volumes or attach existing volumes directly to compute instances.
Retrieve current account quotas and detailed hourly usage reports for billing purposes.
Ask an AI about this
Waiting for input…
Civo (Cloud-native Kubernetes Cloud Provider API) - 33 Tools
These tools let you perform every major infrastructure task, including creating clusters, managing volumes, and checking account usage.
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 Civo (Cloud-native Kubernetes Cloud Provider API) on VinkiusAdd Team Member
Adds a specified user account to an existing team structure within Civo.
Attach Volume
Connects a pre-created block storage volume to a running compute instance for...
Create Cluster
Initiates the process of creating an entirely new Kubernetes cluster within your...
Create Domain Record
Adds or updates specific records (like A or CNAME) to an existing DNS domain.
Create Domain
Registers a completely new DNS domain name under your Civo account.
Create Firewall Rule
Applies a specific rule (like allowing port 80) within an existing firewall setup.
Create Firewall
Sets up a new, dedicated firewall profile to manage network ingress and egress.
Create Instance
Launches and provisions a new virtual machine compute instance with defined sizes.
Create Network
Builds an isolated private network segment to contain specific resources from...
Create Team
Creates a new logical team group for managing multiple users and access rights.
Create Volume
Allocates new block storage space that can later be attached to an instance.
Create Webhook
Sets up a URL endpoint that triggers actions when specific events occur within Civo.
Detach Volume
Disconnects an existing volume from its compute instance, making it available for reuse.
Get Charges
Pulls a detailed hourly usage report showing costs incurred by chargeable resources.
Get Quota
Returns the current status of account resource limits and how much you have used.
List Clusters
Retrieves a list of all Kubernetes clusters currently provisioned in your account.
List Disk Images
Fetches a catalogue of available operating system and base disk images for new...
List Domains
Lists all DNS domains that are currently registered under your account.
List Firewall Rules
Displays every active rule configured within a specific firewall profile.
List Networks
Retrieves an inventory of all existing private network segments.
List Regions
Shows every geographical region where you can deploy infrastructure resources.
List Sizes
Provides a list of all available compute instance sizes (e.g., g3.xsmall, m5.large).
List Ssh Keys
Lists the SSH keys currently stored and associated with your account.
Reboot Instance
Performs a hard restart of a compute instance, simulating pulling the plug and plugging it back in.
Recycle Cluster Node
Restarts an individual node within a Kubernetes cluster to refresh its connection state.
Resize Instance
Increases or decreases the CPU/memory capacity of an existing compute instance.
Retag Instance
Adds or changes descriptive tags on a running instance for organizational purposes.
Soft Reboot Instance
Initiates a graceful software reboot of the compute instance, allowing services to...
Start Instance
Starts an instance that has been previously stopped and is currently offline.
Stop Instance
Shuts down a running compute instance to save costs and halt operations.
Test Webhook
Sends a test payload through an existing webhook endpoint to verify its functionality.
Update Team Member Status
Changes the operational status (active, suspended) of a specific team member account.
Upload Ssh Key
Upload an SSH key
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 Civo 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 every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Civo (Cloud-native Kubernetes Cloud Provider API), then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Civo. 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 connection provides 33 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
The current process for spinning up and securing a new service is tedious.
Right now, deploying a microservice requires jumping between five different consoles. You have to create a network in one tab, write firewall rules in another, provision the instance size on a third, find an available volume on a fourth, and then finally run it all together by clicking 'deploy.' It's slow, error-prone, and you lose track of which rule applied where.
With this MCP, that whole sequence collapses. You tell your agent what you need—say, 'Give me an isolated environment for the billing service.' The agent handles the network setup (`create_network`), applies all necessary security rules (`create_firewall_rule`), provisions the instance (`create_instance`), and attaches storage, giving you a fully ready-to-go resource in minutes.
The Civo MCP gives you total control over infrastructure provisioning.
You eliminate manual steps like visiting the networking section to create the network, then going to the compute section, and manually selecting a volume ID. You just ask for it all together.
It's not about listing services; it's about composing them. Your agent handles the handoffs between tools—from `create_volume` to `attach_volume`—so you only deal with high-level intent.
What your AI can actually do with this
Need to spin up a test environment or scale production resources? This MCP gives you full control over cloud-native infrastructure directly from your agent. You can build out complex microservice architectures by creating dedicated private networks, setting strict firewall rules, and launching multiple compute instances—all without touching the console dashboard.
Think of it like having an expert DevOps engineer sitting next to you who can provision any resource instantly just by reading what you need. If you're using Vinkius, this MCP plugs right into your existing catalog so you don't have to juggle dozens of separate APIs or tools. You manage everything from core K3s clusters and attaching persistent volumes to checking your account usage and billing reports, all conversationally.
019e5d07-3fac-718f-9c04-018101cca965 
Here's how it actually works
The bottom line is: you treat complex cloud operations like sending an email, letting your agent handle the underlying infrastructure logic.
Subscribe to this MCP on Vinkius and provide your Civo API credentials.
Your agent authenticates the connection, making all infrastructure endpoints available via natural language commands.
You issue a request (e.g., 'Spin up two web servers in region X with network Y'), and the agent executes the necessary sequence of API calls.
Who is this actually for?
This MCP is for anyone who has to deal with cloud resource management manually. If you're tired of clicking through five different dashboards just to provision a dev environment or check if you've hit your CPU core limit, this is for you.
Automating cluster provisioning, scaling compute resources, and managing networking rules without leaving the chat interface.
Monitoring resource quotas and running specific checks like getting charges or rebooting instances on demand to stabilize production systems.
Quickly spinning up isolated development instances for testing new features, or attaching volumes containing local data right from their IDE.
What Changes When You Connect
Spin up isolated environments quickly. Instead of manually configuring networking, you can use the agent to create a dedicated private network and apply strict rules using create_firewall and create_firewall_rule for maximum security.
Stay in control of spending. Check your resource limits anytime with get_quota, or review detailed usage costs by calling get_charges. You always know where the money's going.
Handling state changes is easy. Need to scale up? Use resize_instance for a quick capacity boost, or if an instance is acting weird, use reboot_instance instead of logging into SSH just to restart it.
Build resilient architectures. You can deploy new compute instances using create_instance, and ensure their data persists by immediately attaching persistent storage via attach_volume. No more lost state.
Simplify DNS management. Need a new subdomain? Use the agent to first run list_domains to check availability, then use create_domain followed by create_domain_record.
See it in action
Scaling up for peak traffic
The team's primary database server is running on an instance that can't handle the load. Instead of manually logging in and upgrading, the agent runs list_sizes to find a bigger option and uses resize_instance to immediately boost the compute resources.
Debugging a failing service
A developer suspects network bleed-through is causing errors. They ask their agent to list all private networks, then run create_firewall rules around the specific service to isolate and test the problem.
Onboarding a new department
The operations team needs to give Marketing its own siloed resources. They use the agent to execute create_team, followed by creating isolated networks using create_network for that department's specific compute instances.
Deploying a test environment
A developer needs an isolated sandbox to test against production data. They ask the agent to create a new cluster using create_cluster, and then attach a volume created via create_volume for testing purposes.
The honest tradeoffs
Ignoring network segmentation
Just calling 'create an instance' without thought. This leaves the new resource exposed to default, potentially too-broad firewall rules.
Always create a dedicated private network first using create_network. Then use create_firewall and add specific ingress/egress policies with create_firewall_rule before provisioning the compute instance.
Assuming persistence
Running an instance for a test, stopping it, and then starting it again later. All the data is gone because the local disk was ephemeral.
Always plan ahead: use create_volume to provision block storage, and ensure you attach it using attach_volume before the compute instance shuts down.
Over-relying on manual dashboards
Having to switch between the billing console, the networking tab, and the cluster management screen just to check one resource limit.
Ask your agent to run get_quota first. This gives you an immediate overview of usage across instances, network limits, and CPU cores in a single output.
When It Fits, When It Doesn't
Use this MCP if your workflow requires managing interconnected infrastructure components: compute, networking, and persistent storage. If the core task is 'I need to make X talk securely to Y,' use this. Don't use it if you only need to manage a single resource in isolation (e.g., just listing domains). For simple domain management, create_domain works fine on its own; but for anything involving live computing or traffic flow, the comprehensive tooling here is necessary because it lets your agent chain together multiple functions—like creating a network and attaching a volume and creating an instance—which is key to building production-grade systems. If you're just checking prices, get_charges handles that fine, but if you need to do something with the infrastructure data, this MCP has the tools.
Questions you might have
How do I use the create_cluster tool? +
You ask your agent to run create_cluster, specifying the required K3s version and region. The MCP handles the complex, multi-step process of initializing a new Kubernetes environment.
Can I check my account usage using get_quota? +
Yes, running get_quota immediately pulls your current resource utilization against your hard limits. This is essential for preventing unexpected service outages due to over-provisioning.
What's the difference between reboot_instance and soft_reboot_instance? +
Use soft_reboot_instance when you want services to shut down gracefully. Use reboot_instance only if a graceful shutdown fails, as it performs a hard restart.
Do I need create_firewall and create_firewall_rule separately? +
Yes. You must first establish the perimeter by calling create_firewall, and then you use create_firewall_rule to define exactly what traffic is allowed through that new firewall.
When should I use the `attach_volume` tool and how do I safely remove a volume using `detach_volume`? +
You must run detach_volume before attempting to attach it elsewhere. This prevents resource conflicts. The process ensures the volume is unlinked from its original instance first, making it available for reattachment.
What's the recommended procedure and impact when I use the `resize_instance` tool? +
Resizing an instance usually requires a brief maintenance window. The MCP handles the upgrade or scale-up process, but your application should be prepared for potential temporary downtime during resource reallocation.
Before creating a DNS record, how do I check existing domains using `list_domains`? +
You run list_domains to see all registered domain names attached to your account. This lets you confirm the correct namespace and prevents errors when defining new records.
How often should I use the `get_charges` tool to monitor my actual spending? +
You can run get_charges at any time for a real-time usage report. It provides an hourly breakdown of charges, helping you pinpoint exactly which resources are costing money.
Can I create a Kubernetes cluster in a specific region? +
Yes, you can use the create_cluster tool and specify the region parameter (e.g., 'lon1' or 'nyc1') along with the cluster name and network ID.
How do I restart a stuck compute instance? +
You can use the reboot_instance tool for a hard reboot or soft_reboot_instance for a graceful restart by providing the Instance ID.
Can I check my current cloud spending and limits? +
Absolutely. Use the get_charges tool to see your current billing details and get_quota to check your account resource limits.
We've already built the connector for Civo. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 33 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.