MCPSERV.CLUB
kocierik

MCP Nomad Go

MCP Server

Go‑based MCP server that connects to HashiCorp Nomad

Active(78)
38stars
2views
Updated 16 days ago

About

MCP Nomad Go is a lightweight, Golang implementation of the Model Context Protocol that interfaces with HashiCorp Nomad’s HTTP API. It allows tools like Claude and Inspector to query, list, and interact with Nomad jobs directly through MCP.

Capabilities

Resources
Access data sources
Tools
Execute functions
Prompts
Pre-built templates
Sampling
AI model interactions

Demo of Nomad MCP Server in action

The nomad-mcp server is a lightweight, Golang‑based bridge that exposes the full functionality of a HashiCorp Nomad cluster to any Model Context Protocol (MCP)‑compatible AI assistant. By translating Nomad’s HTTP API into MCP resources, tools, and prompts, the server lets developers query cluster state, submit jobs, or stream job logs directly from an AI chat interface. This eliminates the need for manual CLI usage or custom SDKs, enabling rapid prototyping and automation in a conversational workflow.

Key capabilities include:

  • Resource discovery: The server enumerates Nomad allocations, jobs, and nodes as MCP resources that an assistant can browse or filter.
  • Tool integration: Actions such as , , and are exposed as callable tools, allowing the AI to trigger real‑world changes with a simple natural‑language request.
  • Prompt templates: Predefined prompts guide the assistant in constructing valid Nomad API calls, reducing the learning curve for users unfamiliar with Nomad’s JSON schema.
  • Streaming support: Using the transport, log output and job status updates can be streamed in real time, giving the assistant an up‑to‑date view of long‑running tasks.

Typical use cases involve:

  • Continuous deployment: A developer can ask the assistant to “deploy version 2.3 of service X” and have the job submitted automatically, with status updates streamed back.
  • Operational troubleshooting: By querying allocation logs or node metrics through the assistant, engineers can diagnose issues without leaving their chat environment.
  • Infrastructure as code: The assistant can read and modify Nomad job specifications, facilitating dynamic resource scaling or configuration changes driven by conversational commands.

Integration is straightforward for MCP‑aware tools such as Claude Desktop. Once the server is running, it can be added to the assistant’s configuration via Smithery, , or a pre‑built binary. The server listens on a configurable port and supports multiple transport modes (stdio, SSE, or streamable HTTP), ensuring compatibility across diverse client setups.

Overall, nomad-mcp provides a seamless, conversational interface to Nomad, empowering developers to manage workloads and infrastructure through natural language while preserving the power and flexibility of Nomad’s orchestration capabilities.