Tutorial: Sell Your API
You built an API. Now put it on the marketplace and get paid every time someone calls it.
What you need
- An API server — any language, any framework, responding to HTTP requests
- A public IP or domain — your server must be reachable from the internet
- Docker installed somewhere (your server, your laptop, anywhere)
That is it. No SDK to install in your API code. No Opsalis library to import. Your API stays exactly as it is.
Step 1: Build your API server
If you already have an API running, skip to Step 2. Your API can be built in any language — Go, Python, Node.js, Java, Rust, anything. As long as it responds to HTTP requests, it works with Opsalis.
Best practice: include a /health endpoint. It helps you verify your server is running, and the system uses it during validation.
Step 2: Deploy it
Your API server needs to be reachable from the internet. Common options:
- VPS — DigitalOcean, Hetzner, Vultr, OVH ($5-10/month)
- Cloud — AWS EC2, GCP Compute, Azure VM
- Home server — any machine with port forwarding or a tunnel
Verify your API is reachable by testing it from a different network (e.g., your phone on cellular data).
Opsalis does not host your API. Your server runs independently. Opsalis connects paying consumers to it.
Step 3: Set up your control center
Install the control center via Docker (see Installation). Open the web console, create your account, and fund your wallet with a small amount of ETH for transaction fees. $5-10 is enough for months.
Step 4: Register your API
In the web console, go to the API Management tab and click Register API.
- Fill in the form: name, description, your API's URL, protocol, price per call, and OpenAPI spec
- Click Validate API — the system checks your API is reachable and captures demo samples
- Review the preview
- Click Publish
Your API is now live on the marketplace.
Step 5: Monitor sales
The Dashboard in your web console shows real-time stats:
- Calls served — today, this week, all time
- Revenue earned in USDC
- Per-API breakdown if you have multiple APIs
- Response times
Set up alerts so you get an email if your ETH runs low.
Pricing tips
| API type | Suggested price per call |
|---|---|
| Simple lookups (IP geolocation, currency rates) | $0.0001 – $0.001 |
| Standard queries (weather, stock quotes, search) | $0.001 – $0.005 |
| Complex operations (AI inference, image processing) | $0.005 – $0.02 |
| Premium data (financial feeds, proprietary datasets) | $0.01 – $0.10 |
Start low and raise prices as demand grows — you can update pricing at any time from the web console.
All prices are in USDC, a stablecoin worth $1 USD. No crypto volatility — $0.005 today is $0.005 tomorrow.
Common mistakes
1. API not reachable from the internet
The most common issue. Your API must be on a public IP or domain. Test it from a different network before registering.
2. Missing or incomplete OpenAPI spec
Consumers rely on your spec to generate SDKs and understand your endpoints. A good spec means more buyers.
3. Skipping demo samples
Demo samples are the first thing consumers see. Without them, buyers have no way to preview your data format. Include 2-3 samples.
4. API goes down after registration
Use a process manager to keep your API running. Set up monitoring so you know when it goes down.
5. Pricing too high on day one
Price low to attract early users. A $0.001 call that gets 10,000 calls/day earns $10/day. A $0.05 call with zero usage earns nothing.
What happens next
Once your API is live:
- Consumers discover it in the catalog and try the demo for free
- When they switch to production mode, real calls hit your API server
- You get paid per call — USDC appears in your wallet automatically
- The catalog shows your API as "live" so consumers know it is available
Keep your API server running and watch the revenue grow.