# Zeabur PaaS Deployment MCP for AI Agents AI Agent Connect > Zeabur (PaaS Deployment) MCP lets you manage your cloud infrastructure and deployment workflows through your AI agent. It handles service deployments, container commands, and transactional emails directly from your chat interface. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/undefined/ai-agent-connect - **Tags:** paas, deployment, container-management, transactional-email, cloud-hosting ## Description Imagine you're in the middle of a deployment and something breaks. Instead of hunting through multiple tabs to find the right log or trying to remember a complex shell command, you just tell your agent what's wrong. This Connector connects your Zeabur account to your AI client so you can handle the heavy lifting of PaaS management through natural conversation. You can deploy new services using YAML specs, jump into a container to run commands or grab a specific file, and even manage your email infrastructure for transactional messages. It takes the friction out of the ops part of your day. Whether you're spinning up new environments or debugging a failed build, you can do it without switching contexts. It's a way to make your cloud infrastructure feel like an extension of your chat interface. You'll find this works great when you're looking for a unified way to manage resources, which is why it's a staple in the Vinkius catalog for developers who need speed. ## Tools ### create_upload_stage Sets up a place to upload files for your application. This is the first step for pre-packaged deployments. ### deploy_template Launches a new service using a YAML specification. It makes starting new projects much faster. ### download_file Grabs a specific file out of a service container. This helps you inspect files without manual SSH. ### execute_command Runs a shell command inside a container. You can use this to check environments or run scripts. ### get_build_logs Pulls the latest logs from a deployment. It helps you see exactly why a build failed. ### send_email Sends a single transactional email immediately. Use this for things like password resets or receipts. ### prepare_deployment Gets a deployment ready after you've uploaded a file. It handles the final steps needed before launch. ### schedule_email Sets a future time for an email to be sent via Zeabur. This is perfect for delayed notifications. ### send_batch_emails Sends out multiple personalized emails at once. It's the fastest way to handle bulk communications. ## Prompt Examples **Prompt:** ``` Deploy a new service using this YAML: [content] ``` **Response:** ``` I've started the deployment using `deploy_template`. Your new service is being provisioned now. I'll let you know when it's live. ``` **Prompt:** ``` Get the logs for dep-123 ``` **Response:** ``` Fetching logs... The build failed during the 'npm install' phase because the 'express' package wasn't found in your requirements. ``` **Prompt:** ``` Run 'ls -la' in srv-789 ``` **Response:** ``` Running the command... Here are the files in your /app directory: package.json, src, and node_modules. ``` ## Capabilities ### Deploy services from YAML Launch new services using your existing YAML specifications. ### Run container commands Execute shell commands directly inside your service containers. ### Fetch real-time build logs Get immediate access to build output to debug deployment issues. ### Send transactional emails Send single or batch personalized emails via the Zeabur Email API. ### Manage ZIP deployments Handle upload stages and preparation for pre-packaged applications. ### Download container files Grab specific files from your service containers for inspection. ## Use Cases ### Debugging a failed build A developer asks why a deployment failed. The agent uses `get_build_logs` to identify a missing dependency and reports it back immediately. ### Sending a welcome series A backend dev asks the agent to send 500 personalized emails. The agent uses `send_batch_emails` to handle the task in one go. ### Inspecting a production file An engineer needs to see a config file in a running container. The agent uses `download_file` to grab it and display the content. ### Quick service launch A dev provides a YAML template and asks to deploy. The agent uses `deploy_template` to spin up the new service in seconds. ## Benefits - Faster deployments by using `deploy_template` to launch services from your YAML specs. - Instant debugging by pulling logs with `get_build_logs` directly into your chat interface. - Easier file access by using `download_file` to grab data from containers without manual SSH. - Automated communications using `send_batch_emails` for high-volume transactional notifications. - Simplified container access by running scripts with `execute_command` on the fly. - Streamlined app uploads by creating stages with `create_upload_stage` for pre-packaged apps. ## How It Works The bottom line is you get a conversational interface for your entire Zeabur cloud infrastructure. 1. Subscribe to the Connector and provide your Zeabur API Token. 2. Connect your AI client to the Vinkius catalog. 3. Ask your agent to deploy, debug, or send emails. ## Frequently Asked Questions **Can I use Zeabur PaaS Deployment MCP to send emails?** Yes. You can send single transactional emails or handle batch personalized emails directly through your agent. **How does the Zeabur PaaS Deployment MCP handle container commands?** It allows your agent to execute shell commands inside your service containers, making it easy to check environments or run scripts. **Can I deploy new services using YAML with Zeabur PaaS Deployment MCP?** Yes. You can provide a YAML specification to your agent, and it will use the deployment tools to launch the service for you. **Is the Zeabur PaaS Deployment MCP good for debugging?** It's excellent for debugging. Your agent can fetch real-time build logs so you can see exactly why a deployment failed without leaving your chat. **Can I use Zeabur PaaS Deployment MCP to manage my cloud infrastructure?** Yes. It provides a way to manage your Zeabur PaaS resources, including container management and deployment workflows, via natural language. **How do I get files out of my containers using Zeabur PaaS Deployment MCP?** The agent can download specific files from your service containers for you, which is great for inspecting configuration or logs. **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.