skills.rest
Thunder-Compute
Thunder-ComputeOFFICIAL
Explore all skills by this author →
GitHub

Thunder Compute

Create and manage Thunder Compute cloud GPU instances with SSH and port forwarding.

3|2|Updated Sep 22, 2024
Search Tags:#snapshots#mcp tools#instance management#port forwarding#gpu provisioning#billing visibility
One-click install
npx skills add https://github.com/Thunder-Compute/thunder-compute-documentation --skill thunder-compute
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Thunder Compute
Source: https://github.com/Thunder-Compute/thunder-compute-documentation/tree/main
Command: npx skills add https://github.com/Thunder-Compute/thunder-compute-documentation --skill thunder-compute

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of helping you create, connect to, and manage cloud GPU instances without getting stuck on operational details or unsafe lifecycle actions.

Core Features & Use Cases

  • Instance lifecycle management: Create, list, modify (resize/change GPU/mode), snapshot, and delete GPU instances.
  • Operational productivity on the instance: Run commands, set up SSH access, and generate SCP commands for file transfers.
  • Cost and billing visibility: Retrieve metering data and billing/invoice information to understand spend and plan next steps.
  • Port exposure and connectivity: Forward ports so services (e.g., web UIs) become reachable over public HTTPS.

Use case example: Ask your agent to spin up an A100 instance in prototyping mode, wait until it’s ready, run nvidia-smi, forward port 8000, and provide the exact URL and cost impact—then delete the instance when finished.

Quick Start

Use the Thunder Compute skill to create a prototyping-mode GPU instance, then run nvidia-smi and return the instance ID, connection details, and approximate cost.

Frequently Asked Questions about Thunder Compute

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I provision and manage cloud GPU instances for inference workflows?▼

You can provision and manage cloud GPU instances by selecting GPU specs, templates, and prototyping or production modes, then handling the full lifecycle from creation to deletion. The process includes running commands, taking snapshots, forwarding ports, and checking billing visibility.

Can I forward ports to expose a web UI from my GPU instance over public HTTPS?▼

Yes, you can set up port forwarding on your GPU instance to expose services like web UIs over public HTTPS. This connectivity feature makes your development and inference applications reachable externally without manual network configuration.

How do I transfer files to a remote GPU instance using SCP and SSH?▼

You can establish SSH access and generate SCP commands for file transfers to your remote GPU instance. This allows you to move datasets, scripts, and model weights seamlessly during your development or inference workflows.

Does this GPU provisioning approach include guardrails for irreversible deletions?▼

Yes, the GPU instance lifecycle management includes safe intent routing and guardrails for irreversible deletions. It also provides warnings for production-mode billing risks, ensuring safe management and cost control.

How can I check billing invoices and usage metering for cloud GPU instances?▼

You can retrieve metering data and billing invoice information to check usage and spend for cloud GPU instances. This cost and billing visibility helps you understand your consumption and plan your next prototyping or production steps.

Related Skills

Explore More Software Engineering Skills→
AradotsoAradotso
5

terraform-data-engineering-iac

Provision AWS S3 buckets and EC2 instances with Terraform for data engineering.

Official
Advanced
PostHogPostHog
58

workload-analysis

Analyze customer usage and spend to produce multi-view workload dashboards.

Official
Advanced
AirmomoAirmomo

ssh-essentials

Manage SSH keys, tunnels, and file transfers with OpenSSH tools.

Community
Intermediate
KarlinskySKarlinskyS

lambda-labs-gpu-cloud

Provision Lambda Labs GPU cloud instances for machine learning training and inference.

Community
Intermediate