Cloudgpu Discover Compute
Discover and compare GPU compute for an AI agent without an account, and get a quote before any spend. The skill is free; paid CloudGPU compute needs a CloudGPU API key and explicit approval.
Free
Works with the AI Agents you already use
cloudgpu-discover-compute
Example session with this skill installed
I need a GPU with at least 24GB VRAM for a 2-hour training task with a $5 budget. Compare options and get a quote, but do not run anything yet.
- Read your context and instructions
- Compiled the cloudgpu-discover-compute
Summary
The agent read the public CloudGPU GPU catalog (no account needed), compared candidates with at least 24GB VRAM against local compute, and prepared a quote request. Nothing was rented.
Proposal (illustrative format; real values come from the live catalog and quote)
| Item | Value |
|---|---|
| Requirement | min 24GB VRAM, 1 GPU |
| Estimated duration | 2 hours (7200 s) |
| Budget ceiling | $5.00 |
| Candidate GPU / template / disk | taken from the live catalog at quote time |
| Estimated total cost, required starting balance, quote expiry | returned by the quote at request time |
Next steps
- Review the quote and the hard spend/time caps.
- Running paid compute requires a CloudGPU account and API key plus your explicit approval; the free skill itself never spends money.
- If no key is configured, the owner connects an account via the documented connect flow and signs in personally.
Connects securely to your tools. The creator never sees your data.
About this skill
Free skill, paid service
This skill is free. It is instructions only (SKILL.md + MIT LICENSE). It tells your agent how to use the CloudGPU Agent Compute Layer (https://cloudgpu.app).
Important: actually running GPU workloads on CloudGPU is a paid service that requires a CloudGPU account and API key. Reading the public GPU catalog and requesting a quote needs no account and creates no paid resource. Running paid compute requires connecting an account (owner signs in and approves) and an explicit approval of the quoted cost. The skill never silently rents compute, never tops up automatically. Service eligibility: developers outside mainland China only; see https://cloudgpu.app/terms.
What it does
- Reads the public GPU catalog (no account) to compare VRAM, templates and disk options
- Prefers local compute when suitable
- Requests a quote for task-specific requirements and budget
- Presents the proposal and asks for your approval before any paid run
- Keeps API keys in the client's secure credential mechanism
Links
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Instructions only; it does not include a runnable workload. Paid CloudGPU compute requires a CloudGPU account and API key. Service eligibility: developers outside mainland China only.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 9 days ago
- Free with an Agensi account
Nothing to install. It runs in your AI agent.
- Optional
- CloudGPU API key
A CloudGPU account and key are only needed for paid compute.
Permissions
No special permissions declared or detected