Thunder Compute GPU instance management. Create, monitor, and manage cloud GPU instances directly from Claude Code.
Manage Thunder Compute GPU instances directly from your AI coding agent. Create, monitor, modify, and tear down cloud GPU instances without leaving your coding workflow.
/plugin install thunder-compute
Or from GitHub:
/plugin marketplace add Thunder-Compute/thunder-plugin
/plugin install thunder-compute@Thunder-Compute/thunder-plugin
Add the MCP server directly:
codex mcp add thunder-compute --url https://api.thundercompute.com:8443/mcp
Or install the plugin from this repo (place it in your project or ~/.agents/plugins/).
Add Thunder Compute to Cursor's MCP config at ~/.cursor/mcp.json (global) or .cursor/mcp.json (per-project):
{
"mcpServers": {
"thunder-compute": {
"type": "http",
"url": "https://api.thundercompute.com:8443/mcp"
}
}
}
Then open Cursor Settings → MCP and enable thunder-compute. You'll be prompted to sign in via OAuth on first use.
Connect to the streamable HTTP endpoint with OAuth:
https://api.thundercompute.com:8443/mcp
No local installation or API tokens required — authentication is handled via OAuth in your browser.
/tnr/tnr create an a6000 instance with pytorch
/tnr list my instances
/tnr how much have I spent this month?
$tnr$tnr create an a6000 instance with pytorch
$tnr list my instances
$tnr how much have I spent this month?
create an a6000 instance with pytorch
list my instances
delete instance 3
what GPUs are available and how much do they cost?
forward port 8080 on my instance
snapshot my instance before I make changes
run nvidia-smi on my instance
how much have I spent this month?
create an API token for CI
The plugin exposes 28 MCP tools via the remote thunder-compute server:
| Tool | Description |
|---|---|
list_instances | List all GPU instances |
create_instance | Create a new instance |
delete_instance | Delete an instance |
modify_instance | Change instance config |
run_command | Execute a shell command on a running instance |
| Tool | Description |
|---|---|
get_specs | Available GPU specs |
get_availability | Current GPU availability per spec |
get_pricing | Per-hour pricing |
list_templates | OS templates |
| Tool | Description |
|---|---|
list_snapshots | List snapshots |
create_snapshot | Create a snapshot |
delete_snapshot | Delete a snapshot |
| Tool | Description |
|---|---|
list_ssh_keys | List SSH keys |
create_ssh_key | Add an SSH key to your org |
delete_ssh_key | Delete an SSH key |
add_ssh_key_to_instance | Add an SSH key to a running instance |
| Tool | Description |
|---|---|
list_ports | List forwarded ports |
forward_port | Forward HTTP ports |
delete_port | Remove forwarded ports |
| Tool | Description |
|---|---|
get_ssh_command | Get SSH connection command |
get_scp_command | Get file transfer command |
| Tool | Description |
|---|---|
get_meter_data | GPU usage metrics for a time period |
get_upcoming_invoice | Estimated charges for current billing period |
get_invoice_history | Historical invoices |
get_subscription | Subscription plan and payment info |
| Tool | Description |
|---|---|
list_tokens | List API tokens |
create_token | Create an API token |
delete_token | Delete an API token |
MIT
External network access
Connects to servers outside your machine
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub thunder-compute/thunder-plugin --plugin thunder-computeClaude Code skill pack for Vast.ai (24 skills)
SkyPilot agent skill for launching cloud VMs, Kubernetes pods, and Slurm jobs across 25+ clouds
Machine learning training and inference pipeline using cloud GPUs (Modal, Lambda Labs, RunPod) with HuggingFace ecosystem - no local GPU required
Skills for NVIDIAs ecosystem spans GPU acceleration, CUDA, AI agents, inference, robotics, Physical AI, Omniverse, and simulation. This plugin helps you understand the pieces, choose a path, validate your setup, and build practical NVIDIA-powered workflows.
Hetzner Cloud CLI skill for servers, networks, firewalls, load balancers, DNS, volumes, and storage.
Node Hardware MCP - Comprehensive Hardware Monitoring and System Analysis for LLMs with real-time performance metrics