# HashiCorp Nomad MCP for AI Agents AI Agent Connect > HashiCorp Nomad MCP lets you manage your cluster's workloads and orchestration through your AI agent. You can check job statuses, monitor node health, and handle deployments like rollbacks or promotions without jumping between different dashboards. It puts your entire infrastructure under one conversational interface. ## Overview - **Category:** loved-by-devs - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_HSllBsK0lF7lUrIFLAuIb4teHoynOaI9ff1XV1bj/ai-agent-connect - **Tags:** workload-orchestration, container-management, cluster-monitoring, deployment, infrastructure-as-code ## Description This HashiCorp Nomad MCP connects your orchestration cluster to your AI agent so you can manage workloads through a simple conversation. Instead of hunting through the Nomad UI or running manual CLI commands to see why a job is stuck, you can just ask your agent for the status. It handles the heavy lifting of fetching data across your nodes and jobs, making it easy to keep an eye on cluster health or see how a rolling update is progressing. You can use it to see every running allocation, check specific task details, or even manually trigger a deployment promotion when everything looks good. Because Vinkius hosts this Connector, you can plug it into your existing workflow and get immediate visibility into your infrastructure. It takes the friction out of operational tasks by letting you perform deep inspections of metadata or quickly fail underperforming deployments. You get a direct line to your cluster's state without the usual overhead of navigating complex menus. ## Tools ### get_deployment Fetch the specific details and status of a deployment. This is useful for checking the progress of a rolling update. ### get_job Pull the complete configuration and status for a specific job. Use this to see the exact parameters of a running service. ### get_node Get the current status and resource info for a specific client node. This helps you troubleshoot individual server issues. ### list_allocations See a list of all task allocations currently running in your cluster. Use this to get a bird's eye view of active workloads. ### list_deployments See a list of your most recent deployments. This lets you quickly track the history of your recent rollout attempts. ### list_jobs List every job currently registered in your Nomad cluster. Use this to see what's currently scheduled to run. ### list_nodes See a list of all client nodes in your cluster. This is your first stop for checking cluster-wide health. ### promote_deployment Manually promote a successful deployment to the next stage. This is great for moving through canary deployment phases. ### fail_deployment Mark a specific deployment as failed to trigger a rollback. Use this when an update goes wrong and you need to stop it. ### get_allocation Retrieve the specific details for a single task allocation. This helps you see exactly what a specific task is doing. ## Prompt Examples **Prompt:** ``` Show me the status of all the jobs in the production namespace. ``` **Response:** ``` Checking your production jobs now... | Job Name | Status | Type | | :--- | :--- | :--- | | Web-Frontend | Running | Service | | Auth-API | Running | Service | | Worker-Batch | Pending | Batch | | Cache-Node | Running | Service | Would you like me to pull the full configuration for any of these? ``` **Prompt:** ``` Are there any nodes currently down in the cluster? ``` **Response:** ``` I've checked the cluster nodes for you. You have 15 total nodes: - **Healthy**: 13 nodes are currently 'ready'. - **Down**: 2 nodes are currently 'down' or unreachable. I can pull the resource usage for the healthy nodes if you'd like to see the load. ``` **Prompt:** ``` I want to promote the deployment that just finished its canary phase. ``` **Response:** ``` I've identified the most recent successful deployment. I'm initiating the promotion now... **Promotion Status:** - **Deployment ID:** dep-8821 - **Action:** Promoting to all allocations - **Status:** In Progress I'll keep an eye on the rolling update for you. ``` ## Capabilities ### Check job configurations See the full setup and current status of any registered job in your cluster. ### Monitor node health Get real-time status and resource usage for all your client nodes. ### Track deployment progress Follow rolling updates and see the progress of your deployments. ### View running allocations List every active task allocation and inspect specific task details. ### Promote successful deployments Manually move successful deployments forward to the next stage. ### Fail underperforming deployments Mark a deployment as failed to trigger an immediate rollback. ### Fetch job metadata Pull deep metadata for specific jobs, nodes, or allocations using their IDs. ## Use Cases ### Stalled deployment troubleshooting An SRE asks the agent to list_deployments to see where an update stalled, then uses get_deployment to find the error. ### Rapid cluster health checks An admin asks the agent to list_nodes to see if any servers are offline during a heavy traffic spike. ### Canary promotion A developer sees a successful canary and tells the agent to promote_deployment to roll it out to the whole cluster. ### Crash investigation A dev uses get_allocation to see the specific details of a crashed task to find out why it failed. ## Benefits - You get instant visibility into cluster health by using list_nodes to see which nodes are ready or down without manual checks. - You can handle deployment rollbacks faster by using fail_deployment to stop bad updates immediately. - You save time on manual reporting by using list_jobs and list_allocations to get a full snapshot of your workloads. - You can perform deep troubleshooting with get_job to see the exact configuration of any running task. - You manage canary updates more reliably by using promote_deployment to move successful workloads forward. ## How It Works The bottom line is you get a conversational interface for your entire Nomad cluster. 1. Subscribe to the HashiCorp Nomad MCP on Vinkius. 2. Enter your Nomad Address and optional ACL Token in your AI client settings. 3. Ask your agent to list jobs, check node status, or manage deployments. ## Frequently Asked Questions **Can I use the HashiCorp Nomad MCP to check my cluster health?** Yes, it lets you see which nodes are up or down in real time. You can quickly identify hardware issues or connectivity problems without opening a dashboard. **How does the HashiCorp Nomad MCP help with deployments?** It lets you track rolling updates and manually promote or fail them. This makes managing canary releases and rollbacks much faster and less error-prone. **Does the HashiCorp Nomad MCP support multi-region clusters?** It connects to your specific Nomad address to pull data from your cluster. Whether that cluster is local or spread across regions depends on your own setup. **Can I use this to see my job configurations?** Yes, you can fetch the full configuration for any registered job. This is great for verifying that your agent-managed updates actually applied the correct settings. **Is the HashiCorp Nomad MCP good for SRE teams?** It's built for SREs who need to quickly check statuses without opening the UI. It provides a fast way to get the facts during an incident. **Can I use the HashiCorp Nomad MCP to manage my workloads?** Yes, you can list all running allocations and inspect task details. It gives you a way to interact with your active workloads using natural language. **How do I get my Nomad Address?** Your Nomad Address is the URL where your Nomad server is reachable, usually including the port (e.g., http://nomad.example.com:4646). **Is an ACL Token required?** Only if ACLs are enabled on your Nomad cluster. If enabled, you must provide a Secret ID with sufficient permissions to query the requested resources. **Can I use namespaces with this server?** Yes! You can specify a target namespace in the authentication settings. The server will then filter all job and allocation queries to that specific namespace.