Agents and Keys

This project supports running multiple agent identities (private keys) to place orders on Hyperliquid. Agents are lightweight keys authorized on-chain by your main wallet.

Agent storage and environment

- For security, agent private keys should be stored as environment variables and not checked into source control.
- Supported patterns:
  - `AGENT_PRIVATE_KEY` (single default agent)
  - `<NAME>_AGENT_PK` or `<NAME>_AGENT_PRIVATE_KEY` (per-agent keys)

Discovering agents

- `trade_executor.py` scans environment variables for agent keys and loads them into `Exchange` objects so each agent can sign orders independently.

Creating and authorizing agents

- Use `create_agent.py` with your `MAIN_WALLET_PRIVATE_KEY` to authorize a new agent name. The script will attempt to call `exchange.approve_agent(agent_name)` and print the returned agent private key.

Security notes

- Never commit private keys to Git. Keep them in a secure secrets store or local `.env` file excluded from version control.
- Rotate keys if they are ever exposed and re-authorize agents using your main wallet.

Example `.env` snippet

MAIN_WALLET_PRIVATE_KEY=<your-main-wallet-private-key>
MAIN_WALLET_ADDRESS=<your-main-wallet-address>
AGENT_PRIVATE_KEY=<agent-private-key>
EXECUTOR_SCALPER_AGENT_PK=<agent-private-key-for-scalper>

File `agents`

- This repository may contain a local `agents` file used as a quick snapshot; treat it as insecure and remove it from the repo or add it to `.gitignore` if it contains secrets.