Cloud World Model on Smithery
    Canvas Cloud AI Users
    2 minutes to set up

    Unlock full API access

    You've seen what the simulator can do. Get your API key and continue with unlimited steps, programmatic access, RL training environments, and multi-cloud analysis — no credit card required.

    What you unlock with a full API key

    Unlimited simulation steps — no 60-step demo cap
    Programmatic API access for scripts and AI agents
    RL training environments for reward optimization
    Saved scenarios and multi-cloud comparison jobs
    Step 01
    Get your Canvas Cloud AI API key

    Log in to Canvas Cloud AI, go to Account Settings → API Keys, and generate a key. It will start with cca_ and is shown only once — copy it immediately.

    Bash
    # Go to Canvas Cloud AI → Account Settings → API Keys
    # Generate a new key — it starts with cca_
    # Example: cca_live_abc123def456...
    Step 02
    Exchange it for a World Model key

    Paste your Canvas Cloud AI key below. The World Model will verify your identity and return a permanent key scoped to this API.

    Step 03
    Start calling the API

    Use your World Model key (cwm_live_...) in the Authorization header for every request. You now have access to simulations, chaos engineering, multi-cloud analysis, and more.

    curl -X POST https://clouddigitaltwin.com/api/simulations \
    -H "Authorization: Bearer cwm_live_your_world_model_key" \
    -H "Content-Type: application/json" \
    -d '{
    "name": "My First Simulation",
    "provider": "aws",
    "region": "us-east-1"
    }'

    Good to know

    Your key, your scopes

    World Model keys issued via Canvas Cloud AI start with read and write access. Visit the Admin Key Management page (/admin/keys) if you need admin-scoped key setup.

    One Canvas Cloud AI key → many World Model keys

    You can register multiple times with the same Canvas Cloud AI key to create separate keys for different projects or agents.

    Email verification required

    Your Canvas Cloud AI account email must be verified before you can get a World Model key. Check your Canvas Cloud AI account if you see a 403 error.

    Rate limits

    New keys start at 1,000 requests per hour. The limit is tied to your World Model key, not your Canvas Cloud AI key.

    Credit costs

    Your account starts with 1,000 free credits per month — refreshed automatically every 30 days. Status and read endpoints are always free.

    Call typeCredits
    Simulation step1
    RL episode step1
    Chaos job5
    Multi-cloud strategy job5
    AI explanation10
    Status / read endpoints
    Free

    Need more? Buy a credit pack — Starter $9 / 10k, Professional $49 / 100k, Enterprise $299 / 1M. Paid credits never expire.

    Pay per request — no API key needed

    Agents and scripts can pay in USDC on Base via the x402 protocol — no account or API key required.

    TypeScript quick-start
    x402 · USDC on Base
    Bash
    npm install x402-fetch viem
    TypeScript
    import { wrapFetchWithPayment } from "x402-fetch";
    import { createWalletClient, http } from "viem";
    import { base } from "viem/chains";
    import { privateKeyToAccount } from "viem/accounts";
    const account = privateKeyToAccount(
    process.env.AGENT_WALLET_PRIVATE_KEY as `0x${string}`
    );
    const walletClient = createWalletClient({ account, chain: base, transport: http() });
    // Wrap fetch once — every 402 response is paid and retried automatically.
    const fetch402 = wrapFetchWithPayment(fetch, walletClient);
    const res = await fetch402(
    "https://www.cloudworldmodel.ai/api/simulations/sim_abc/step-hybrid",
    { method: "POST", body: JSON.stringify({ steps: 1 }) }
    );
    const { metrics, settlement } = await res.json();
    // metrics.cpuUsage, metrics.costPerHour, settlement.transactionHash

    Python, Go, and cURL examples — plus wallet setup, per-call pricing, and replay-protection details — are in the Pay-per-request section on the Agents page.

    Frequently asked questions

    Verify Simulation Accuracy

    Before you build on the numbers, see how they hold up. The benchmark report runs the simulation engine head-to-head across AWS, GCP, Azure, and OCI — comparing cost accuracy, latency profiles, throughput, and error-rate behavior. Every result is reproducible and re-validated in CI.

    Open Benchmark Report

    Building an AI Agent?

    For AI Agents

    The Cloud World Model merchant page on TryPoncho is the machine-readable agent storefront — listing all x402-priced endpoints, live pricing, the curated OpenAPI spec, and AgentCash integration. No account needed; pay per call in USDC.

    Open Agent Storefront on TryPoncho

    Ready to go deeper?

    The full API reference covers simulations, chaos engineering, RL training, multi-cloud analysis, and more.