PluginBench
MCP Server
Active
Apache-2.0

Terradev MCP Server

io.github.theoddden/terradev

Cross-cloud GPU orchestration and resource management from the CLI.

RepositoryRegistry

What is the Terradev MCP server?

Terradev is a cross-cloud GPU orchestration MCP server that enables CLI-based management of GPU resources across multiple cloud providers. It allows you to provision, monitor, and orchestrate GPU workloads without vendor lock-in.

Terradev provides command-line tools for orchestrating GPU compute resources across different cloud platforms. Use it to manage GPU allocation, job scheduling, and resource optimization in a cloud-agnostic way.

How to install Terradev

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
Environment / auth
  • RUNPOD_API_KEY
    secret

    RunPod API key — get at runpod.io/console/settings/api-keys

  • VAST_API_KEY
    secret

    Vast.ai API key — get at console.vast.ai/api-keys

  • LAMBDA_API_KEY
    secret

    Lambda Labs API key — get at cloud.lambdalabs.com/api-keys

  • HF_TOKEN
    secret

    HuggingFace API token — get at huggingface.co/settings/tokens

  • AWS_ACCESS_KEY_ID
    secret

    AWS access key ID (pair with AWS_SECRET_ACCESS_KEY)

  • AWS_SECRET_ACCESS_KEY
    secret

    AWS secret access key

  • GOOGLE_APPLICATION_CREDENTIALS

    Path to GCP service account JSON key file

  • TERRADEV_MCP_BEARER_TOKEN
    secret

    Bearer token for SSE/remote transport mode (stdio mode does not require this)

~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "terradev": {
      "command": "uvx",
      "args": [
        "terradev-cli",
        "mcp",
        "serve"
      ],
      "env": {
        "RUNPOD_API_KEY": "<YOUR_RUNPOD_API_KEY>",
        "VAST_API_KEY": "<YOUR_VAST_API_KEY>",
        "LAMBDA_API_KEY": "<YOUR_LAMBDA_API_KEY>",
        "HF_TOKEN": "<YOUR_HF_TOKEN>",
        "AWS_ACCESS_KEY_ID": "<YOUR_AWS_ACCESS_KEY_ID>",
        "AWS_SECRET_ACCESS_KEY": "<YOUR_AWS_SECRET_ACCESS_KEY>",
        "GOOGLE_APPLICATION_CREDENTIALS": "<YOUR_GOOGLE_APPLICATION_CREDENTIALS>",
        "TERRADEV_MCP_BEARER_TOKEN": "<YOUR_TERRADEV_MCP_BEARER_TOKEN>"
      }
    }
  }
}

Use cases

  • Provision GPU resources across multiple cloud providers from a single interface
  • Monitor and manage GPU workload distribution and utilization
  • Optimize GPU resource allocation and cost across cloud platforms
  • Schedule and orchestrate distributed GPU compute jobs
  • Automate GPU infrastructure provisioning and scaling

Terradev MCP server FAQ

What is Terradev?

Terradev is a cross-cloud GPU orchestration CLI that lets you manage GPU resources across multiple cloud providers without vendor lock-in.

Is Terradev free?

The README does not specify pricing details. Check https://github.com/theoddden/Terradev for current information.

How do I install Terradev?

Install via PyPI with `pip install terradev-cli` or connect remotely via the SSE endpoint at https://terradev-mcp.terradev.cloud/sse.

What cloud providers does it support?

The README does not list specific supported cloud providers. Refer to the repository documentation for details.

Do I need authentication?

The README does not specify authentication requirements. Check the repository for setup and credential configuration details.

Can I use this in Cursor or Claude?

Yes, you can configure it as an MCP server in Cursor or Claude using the PyPI package or the remote SSE endpoint.

Related MCP servers

IOio.github.theoddden/atomic-clock logo

io.github.theoddden/atomic-clock

Current UTC time from NTP atomic clocks. Zero-dependency stdio server.

View repository →
IOio.github.theoddden/stamp logo

io.github.theoddden/stamp

Concierge MCP for agentic workflows: verified time + drift, uuid, diff, calc, attest, verify.

View repository →
FOForm4API — SEC Insider Trading logo

Form4API — SEC Insider Trading

Active

Real-time SEC Form 4, Form 144, 13F-HR & congressional STOCK Act insider data. 37 tools + 6 prompts.

1
TypeScript
MIT
View repository →
HUHumanOps logo

HumanOps

MCP server for AI agents to dispatch real-world tasks to verified human operators

View repository →
IOio.github.theprodsde/agent-memory logo

io.github.theprodsde/agent-memory

Active

Persistent agent memory with a decision layer: replay, restore, verify, or none — explainable.

0
Python
View repository →
IOio.github.therawlogs/kineti logo

io.github.therawlogs/kineti

Active

Kineti: spend cap, undo, and test proof for AI coding agents

0
Rust
MIT
View repository →