This guide takes you from zero to a working Rivano setup: gateway running locally, your first request proxied, and a trace visible in the dashboard.

Prerequisites:

Node.js 18+ installed
An OpenAI API key (or any supported provider)
A Rivano account at app.rivano.ai

Step 1 — Get your API key

Log in to app.rivano.ai and navigate to Settings → API Keys. Create a new key with api scope and copy it — the full value is only shown once.

Step 2 — Install the CLI and gateway

npm install -g @rivano/cli @rivano/gateway

Authenticate the CLI with your API key:

rivano login
# Rivano API key: rv_...
# Logged in as you@example.com

Step 3 — Create your first agent

rivano init

This creates a rivano.yaml in the current directory. Open it and define an agent:

agents:
  - name: contract-summarizer
    description: Summarizes legal contracts
    modelProvider: openai
    modelName: gpt-4o
    environment: production

Deploy the agent to the control plane:

rivano deploy
# ✔ Validated rivano.yaml
# ✔ Deployed agent: contract-summarizer

Step 4 — Start the gateway

Add your provider key to rivano.yaml:

gateway:
  port: 8080
  providers:
    openai:
      apiKey: "sk-..."
  cloud:
    apiKey: "rv_..."
    baseUrl: "https://api.rivano.ai"

Start the gateway:

rivano-gateway start
# Gateway running on http://localhost:8080
# Connected to Rivano control plane

Step 5 — Send your first proxied request

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
  baseURL: 'http://localhost:8080/openai/v1',
});

const response = await client.chat.completions.create({
  model: 'gpt-4o',
  messages: [
    {
      role: 'system',
      content: 'You are a legal contract summarizer. Be concise.',
    },
    {
      role: 'user',
      content: 'Summarize: "Party A agrees to deliver 100 units by March 31..."',
    },
  ],
});

console.log(response.choices[0].message.content);

import openai

client = openai.OpenAI(
    api_key="sk-...",
    base_url="http://localhost:8080/openai/v1",
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {
            "role": "system",
            "content": "You are a legal contract summarizer. Be concise.",
        },
        {
            "role": "user",
            "content": 'Summarize: "Party A agrees to deliver 100 units by March 31..."',
        },
    ],
)

print(response.choices[0].message.content)

curl http://localhost:8080/openai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-..." \
  -d '{
    "model": "gpt-4o",
    "messages": [
      {"role": "system", "content": "You are a legal contract summarizer. Be concise."},
      {"role": "user", "content": "Summarize: Party A agrees to deliver 100 units by March 31..."}
    ]
  }'

Step 6 — View the trace

Open app.rivano.ai and go to Traces. Your request appears within seconds, showing:

Model, provider, latency, and token counts
Cost estimate for the request
Policy evaluations (all passed, since you haven’t configured any yet)
Full request and response payloads

💡

Run rivano traces list in your terminal to see the same data without opening the dashboard.

Next steps

Core Concepts — Understand agents, policies, and providers in depth
SDK Overview — Manage Rivano programmatically from your application
Gateway Configuration — Configure middleware, rate limits, and multi-provider routing
Policies — Add your first governance rule

Installation — Full install reference with all options
CLI Deploy — All deploy flags and rivano.yaml schema
Gateway Overview — When to run the gateway vs. using the managed endpoint

Quickstart