> For the complete documentation index, see [llms.txt](https://clova-2.gitbook.io/clova-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://clova-2.gitbook.io/clova-docs/operations/deployment.md).

# Deployment Guide

## Architecture

```mermaid
flowchart LR
  U["User browser"] --> V["Vercel - Next.js frontend"]
  V -->|REST + delegation| R["Railway - Node agent + cron"]
  R --> ONE["1Shot relayer"]
  R --> VE["Venice API"]
  R --> PY["Pyth Entropy"]
  ONE --> B[("Base Mainnet - ClovaSavingsPool, adapters, RotationHelper")]
  R -.->|reads/writes| B
```

***

## Deployed Contract Addresses (Base Mainnet)

| Contract                 | Address                                      |
| ------------------------ | -------------------------------------------- |
| ClovaSavingsPool (proxy) | `0x96246c2d585D423931c00703Fa74589458B6050b` |
| AaveAdapter              | `0xac8AA12d6E0Fd04D6A9A726A68af0D1109122557` |
| CompoundAdapter          | `0x727B2c11A70d1C215b7A7455245731694e70AFb9` |
| MoonwellAdapter          | `0xc7D3b8cda22fcD1B0d3d651326E04C46b2A37ba9` |
| RotationHelper           | `0xEa448dF1052212F1E7463628F98e836893DD23E2` |

***

## Smart Contract Deploy (if redeploying)

```bash
cd smart-contract

# Set env vars
cp .env.example .env
# Fill: PRIVATE_KEY, TREASURY_ADDRESS, AGENT_ADDRESS

# Deploy + verify
source .env && forge script script/Deploy.s.sol \
  --rpc-url $BASE_MAINNET_RPC_URL \
  --private-key $PRIVATE_KEY \
  --broadcast \
  --verify \
  --etherscan-api-key $BASESCAN_API_KEY \
  --slow \
  -vvv
```

Output will show new contract addresses to update in backend/frontend env vars.

***

## Backend Deploy (Railway)

### 1. Install Railway CLI

```bash
npm install -g @railway/cli
railway login
```

### 2. Initialize project

```bash
cd backend
railway init   # Create new project → name: clova-agent
```

### 3. Set environment variables

```bash
railway variables set \
  AGENT_PRIVATE_KEY=0x... \
  POOL_ADDRESS=0x96246c2d585D423931c00703Fa74589458B6050b \
  ADAPTER_ADDRESS=0xac8AA12d6E0Fd04D6A9A726A68af0D1109122557 \
  COMPOUND_ADAPTER_ADDRESS=0x727B2c11A70d1C215b7A7455245731694e70AFb9 \
  MOONWELL_ADAPTER_ADDRESS=0xc7D3b8cda22fcD1B0d3d651326E04C46b2A37ba9 \
  ROTATION_HELPER_ADDRESS=0xEa448dF1052212F1E7463628F98e836893DD23E2 \
  TREASURY_ADDRESS=0x92F9778c18D43b9721E58A7a634cb65eeA80661d \
  DEPLOYER_ADDRESS=0x92F9778c18D43b9721E58A7a634cb65eeA80661d \
  BASE_RPC_URL=https://mainnet.base.org \
  CHAIN_ID=8453 \
  VENICE_API_KEY=... \
  LIFI_API_KEY=... \
  ONESHOT_RPC_URL=https://relayer.1shotapi.com/relayers \
  ONESHOT_API_KEY=... \
  ADMIN_SECRET=... \
  ROUND_CRON="0 0 * * *" \
  DRAW_CRON="0 0 * * 0" \
  PORT=3001
```

### 4. Deploy

```bash
railway up
```

### 5. Get Railway URL and set webhook

```bash
railway domain   # → https://clova-agent-xxx.railway.app

railway variables set \
  WEBHOOK_BASE_URL=https://clova-agent-xxx.railway.app
```

***

## Frontend Deploy (Vercel)

### 1. Set environment variables in Vercel dashboard

| Key                                   | Value                                        |
| ------------------------------------- | -------------------------------------------- |
| `NEXT_PUBLIC_BACKEND_URL`             | `https://clova-agent-xxx.railway.app`        |
| `NEXT_PUBLIC_POOL_ADDRESS`            | `0x96246c2d585D423931c00703Fa74589458B6050b` |
| `NEXT_PUBLIC_ADAPTER_ADDRESS`         | `0xac8AA12d6E0Fd04D6A9A726A68af0D1109122557` |
| `NEXT_PUBLIC_ROTATION_HELPER_ADDRESS` | `0xEa448dF1052212F1E7463628F98e836893DD23E2` |
| `NEXT_PUBLIC_USDC_ADDRESS`            | `0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913` |
| `NEXT_PUBLIC_AGENT_ADDRESS`           | `0x817E6370fdacA82DEcdD05AD8f464981d05935d3` |
| `NEXT_PUBLIC_CHAIN_ID`                | `8453`                                       |
| `NEXT_PUBLIC_DEPLOYER_ADDRESS`        | `0x92F9778c18D43b9721E58A7a634cb65eeA80661d` |

### 2. Deploy from GitHub

Connect repo to Vercel → auto-deploy on push to main.

Or manually:

```bash
cd frontend
npx vercel --prod
```

***

## Wallet Funding Requirements

| Wallet            | Address                                      | Required              |
| ----------------- | -------------------------------------------- | --------------------- |
| **Agent**         | `0x817E6370fdacA82DEcdD05AD8f464981d05935d3` | 0.001 ETH + 2 USDC    |
| **Pool contract** | `0x96246c2d585D423931c00703Fa74589458B6050b` | 0.001 ETH (Pyth fees) |

***

## Manual Trigger (Demo)

### Via curl (from terminal)

```bash
# Sweep (collect yield from all users)
curl -X POST https://clova-agent-xxx.railway.app/admin/run-cycle \
  -H "x-admin-secret: YOUR_ADMIN_SECRET"

# Draw (pick winner via Pyth VRF)
curl -X POST https://clova-agent-xxx.railway.app/admin/draw \
  -H "x-admin-secret: YOUR_ADMIN_SECRET"
```

### Via wallet signature (from frontend admin panel)

Connect with deployer wallet → admin panel appears → click "Trigger Draw" → sign message in MetaMask.

***

## Verify Deployment

```bash
# Check agent is alive
curl https://clova-agent-xxx.railway.app/status

# Check AI decisions are being logged
curl https://clova-agent-xxx.railway.app/decisions

# Check agent wallet balance
curl https://clova-agent-xxx.railway.app/admin/agent-balance \
  -H "x-admin-secret: YOUR_ADMIN_SECRET"

# Test Venice connection
curl -X POST https://clova-agent-xxx.railway.app/admin/test-venice \
  -H "x-admin-secret: YOUR_ADMIN_SECRET"
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://clova-2.gitbook.io/clova-docs/operations/deployment.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.