Zeabur PaaS MCP for AI. Manage deployments and send emails via natural conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Zeabur PaaS Deployment lets your AI agent manage cloud services, run container commands, and send emails directly from natural conversation.
Deploy full application stacks using YAML templates or pre-packaged ZIP files, download specific runtime logs, and handle complex transactional email campaigns—all without logging into a dashboard.
What AI agents can do with Zeabur (PaaS Deployment) Automation
Create upload stage
Establishes a temporary upload area required for deploying pre-packaged application files.
Deploy template
Deploys an entire service using raw YAML specification files.
Download file
Retrieves a specific file from any active service container.
Pushes new services live using raw YAML specifications, bypassing manual console input.
Executes shell commands within a running service container to check status or modify data.
Pulls specific files out of an active service container for local inspection or modification.
Manages the full process for large application uploads, from creating staging areas to final deployment preparation.
Sends immediate or scheduled batch emails using personalized content and specific API endpoints.
Ask an AI about this
Waiting for input…
What AI agents can do with Zeabur (PaaS Deployment) - 9 Tools
These tools give your agent full control over deploying services, executing commands in containers, and managing all transactional email flows.
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 Zeabur (PaaS Deployment) on VinkiusCreate Upload Stage
Establishes a temporary upload area required for deploying pre-packaged application files.
Deploy Template
Deploys an entire service using raw YAML specification files.
Download File
Retrieves a specific file from any active service container.
Execute Command
Runs arbitrary shell commands inside a live service container environment.
Get Build Logs
Fetches the real-time build output and logs for any given deployment attempt.
Prepare Deployment
Completes the setup phase after a file upload, making the application ready for actual deployment.
Schedule Email
Sets up an email to be sent at a specific time in the future.
Send Batch Emails
Sends multiple personalized emails to a group of recipients simultaneously.
Send Email
Dispatches a single, personalized transactional email immediately.
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 Zeabur PaaS 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 Zeabur (PaaS Deployment), 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 Zeabur. 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
Built on the Model Context Protocol (MCP) for 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 9 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Dealing with cloud infrastructure updates means context switching., Solved with Vinkius AI Gateway
Today, updating a service requires bouncing between your IDE, the web console to initiate deployment, another dashboard to check logs, and finally opening a separate email client to notify stakeholders. You copy IDs, you click 'Next', and you paste error codes into five different places.
With this MCP, you tell your agent exactly what needs doing—'Deploy v2 of service X and let the team know.' The agent handles the deployment using `deploy_template`, pulls necessary logs via `get_build_logs` if anything goes wrong, and wraps it all up by sending a status update email. You get back control of your time.
Sending emails gets automated with send_batch_emails.
The manual process involves setting up mailing lists, writing unique content for every segment, and manually scheduling or sending each communication piece one by one. It's slow, and it’s prone to human error when dealing with large numbers of recipients.
Now, you tell the agent to run `send_batch_emails`. You provide a list of contacts and the template content. The system handles the personalization loop, ensuring every recipient gets a unique message at scale. It's done.
What your AI can actually do with this
You can use this MCP to automate the entire lifecycle of your cloud infrastructure, treating deployment and communication like simple conversations with your agent. Need to deploy a new backend service? Just give the YAML template, and the agent handles the whole process. Want to debug an issue? The agent fetches real-time build logs so you don't have to leave your chat window.
You can run arbitrary shell commands inside a running container or download specific files for inspection. Beyond deployments, it handles all email traffic: sending single messages, scheduling future sends, and managing batch personalized emails. Because these operations involve sensitive credentials and critical system calls, Vinkius enforces the use of a zero-trust proxy.
This means your API keys pass through only in transit; they never sit on disk. You just focus on what needs to happen.
019e5d69-3718-73c8-bc05-b1fdc61dd072 
Here's how it actually works
The bottom line is you manage complex cloud tasks—deploying code, running commands, sending emails—using only natural language prompts.
Subscribe to this MCP and provide your Zeabur API Token (and email token if needed).
Tell your agent the goal, like 'Deploy service X using this YAML' or 'Send a reminder batch for Q3'.
The agent executes the required steps through the necessary tools and reports back status updates.
Who is this actually for?
DevOps engineers and full-stack developers who are tired of context switching between multiple web consoles to deploy a service or send an email blast. This MCP lets you manage the whole pipeline from one place.
Automates service scaling, runs diagnostic commands inside containers, and ensures environment parity across stages.
Deploys new features using YAML templates and checks build logs without leaving their primary development tool.
What Changes When You Connect
Deploy services using deploy_template, which lets you manage entire backend stacks just by providing a YAML file, cutting out the manual step-by-step console setup.
When debugging code, use get_build_logs to pull real-time build output directly into your chat. You don't need to switch tabs or copy/paste error chunks.
execute_command lets you run shell commands inside a running container. This is critical for quick diagnostics when the application isn't throwing clear errors.
Managing email campaigns becomes simple. Instead of writing code, use send_batch_emails to send personalized updates to hundreds of contacts at once.
The full deployment process is covered. Start with create_upload_stage, then use prepare_deployment before the final push, giving you control over every step.
It handles scheduling too. If an email needs to go out next Tuesday morning, just ask your agent to run schedule_email.
See it in action
Rolling back a broken feature flag
A developer runs into production errors and asks the agent to check the logs. The agent uses get_build_logs, determines the last stable state, and initiates a targeted rollback via deploy_template.
Sending quarterly status reports
The marketing team needs to notify 50 clients about new features. The agent uses send_batch_emails, pulling personalized data for each recipient and sending the full communication without manual CSV uploads.
Investigating a container leak
An ops engineer suspects memory usage is spiking in a service container. They instruct the agent to run execute_command (e.g., 'top -b') and get the live output for immediate diagnosis.
Preparing assets for a new app version
A front-end team needs to deploy their latest ZIP package. They first use create_upload_stage, then prepare_deployment before calling deploy_template to ensure the service is fully ready.
The honest tradeoffs
Manually scripting every command
The user writes out a long script containing dozens of shell commands and fails because one syntax error breaks the whole process.
Let your agent orchestrate it. Instead of writing complex scripts, use execute_command to run single, isolated diagnostic checks or combine calls like download_file followed by execute_command to complete a task.
Mixing deployment logic with email sends
Trying to write conditional code that says: 'If deploy fails, send an error email.' This requires complex branching logic outside the scope of simple tool calls.
Build your workflow in stages. First, use deploy_template for the service update. Then, if needed, follow up with a separate request to send_email to notify stakeholders about the outcome.
Assuming all files are available
Trying to run diagnostics on a file that hasn't been uploaded yet, resulting in 'File not found' errors.
Always start by calling create_upload_stage and then use prepare_deployment. This ensures the environment is fully set up before attempting any actions like download_file.
When It Fits, When It Doesn't
Use this MCP if your workflow requires a combination of infrastructure management (YAML deployments, container commands) AND outbound communication (scheduled or batch emails). Don't use it if you only need to send messages; another messaging tool is better. Also, don't use it just for monitoring—if you only want logs, running get_build_logs is sufficient. If your core process involves complex data transformation before deployment, focus on that logic first, then let this MCP handle the execution and notification parts.
Questions you might have
How do I use send_email with this MCP? +
You simply ask your agent to send an email and provide the necessary details, like recipient list and subject line. The tool handles the API call, making it a single-step action.
Can I run multiple commands in one go using execute_command? +
Yes, you can chain sequential commands by listing them as arguments to execute_command. However, be careful: if any command fails, the entire sequence stops. Keep diagnostics simple.
Is there a better way than using deploy_template? +
deploy_template is for structured deployments based on YAML specs. If you have a fully built application in a ZIP file, start by calling create_upload_stage, then use the full lifecycle via prepare_deployment.
What if I need to check logs after deployment? +
Use get_build_logs. You just need to reference the specific deployment ID or service name, and the agent pulls the entire build output history for you. It's a reliable way to confirm success.
How does the MCP handle sensitive credentials when I use a tool like `deploy_template`? +
Your keys pass through a zero-trust proxy. They are only used while transferring data; they never sit on disk. This keeps your actual API tokens safe and contained throughout the process.
What is the correct sequence of calls when I need to deploy a pre-packaged application using `create_upload_stage`? +
You must first execute create_upload_stage to establish the necessary deployment environment. After that, you pass the unique stage ID it returns into prepare_deployment.
If I use `download_file`, what format is the retrieved content in? +
The output is provided as a raw binary stream. You'll get the actual file contents from inside the service container, which your agent can then save or process.
What’s the difference between using `send_email` and `schedule_email`? +
send_email sends a transactional message immediately through the Zeabur API. Conversely, schedule_email queues that message to send at a specific future time or date.
Can I run shell commands inside my running services? +
Yes! Use the execute_command tool by providing the Service ID, Environment ID, and the command array (e.g., ["ls", "-la"]). Your agent will return the output from the container.
How do I debug a failed deployment using this server? +
You can use the get_build_logs tool with the Project ID and Deployment ID. It will fetch the logs so your AI can analyze the errors and suggest fixes.
Does this support deploying pre-packaged ZIP files? +
Yes. First, use create_upload_stage to get a presigned URL and upload ID. After uploading your file, use prepare_deployment to trigger the actual deployment process.
We've already built the connector for Zeabur PaaS. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 9 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.