Agent Layer Node Setup Guide

Follow this guide to download the node bootstrap script, gather your credentials, and start your Agent Layer node locally.

Keep your manual.config.json private. It contains credentials required to run the node.

At a glance

Step 1

Download setup.sh

Run the bootstrap script from the folder where you downloaded it.

Step 2

Fill credentials

Provide wallet, RPC, API key, and model details when prompted.

Step 3

Start the server

The script launches bun dev and keeps the node running on port 8080.

Prerequisites

Before running the setup script, install these dependencies:

Git — Download Git
Bun v1.3.6 — Install with:

curl -fsSL https://bun.sh/install | bash -s "bun-v1.3.6"

Verify with:

bun --version

Ollama — Download Ollama for local model execution

1. Download and run `setup.sh`

Download setup.sh from agentlayerai.vercel.app
Open a terminal in the folder where you downloaded it
Make it executable and run it:

chmod +x setup.sh
./setup.sh

The script will:

Clone the node_server into a local Agent_Layer/ directory
Write default.config.json automatically
Prompt you for the values that belong in manual.config.json

2. Gather your credentials

Wallet

`PRIVATE_KEY` and `PUBLIC_ADDRESS`

Use a wallet on Base Mainnet, such as MetaMask. Your public address is the 0x... address shown in the wallet, and your private key comes from the wallet settings.

Never share your private key. Keep manual.config.json out of version control.

RPC

`BASE_RPC_URL`

Create a Base Mainnet RPC endpoint in Alchemy and copy the HTTPS URL into this field.

https://base-mainnet.g.alchemy.com/v2/your-api-key

Endpoint

`NODE_QUERY_URL`

This value is fixed and should remain set to the Agent Layer query endpoint below.

https://agentlayerai.vercel.app/

API key

`NODE_API_KEY`

`MODEL` - Ollama model name

Set MODEL to any Ollama model you want to run locally, based on the hardware available on your machine. Pull the model first, then use the same model name in the config.

ollama pull <your-model-name>

For model selection, installation help, and usage details, refer to the Ollama documentation.

3. Server starts automatically

Once all values are entered, the script will start the node server for you:

bun install
bun dev

The server should come online on port 8080. Keep that terminal open; closing it stops the node.

To restart it later:

cd Agent_Layer
bun install
bun dev

What gets created

After the script finishes, your Agent_Layer/ directory will look like this:

Agent_Layer/
├── default.config.json
├── manual.config.json
└── ... node_server files

Troubleshooting

Permission denied

Run chmod +x setup.sh and try again.

Git not found

Install Git from the official download page and reopen your terminal.

Bun missing after install

Restart the terminal or reload your shell environment.

Ollama model not running

Make sure Ollama is running in the background, then start your node again.

ollama serve

Support

For help, visit agentlayerai.vercel.app or open an issue on the repository.

Agent Layer Node