Configure authentication

Load your practice-account credentials, then read an account summary to verify access. This tutorial makes a read-only request.

Obtain a token and account ID

Use an OANDA account with v20 API access. In the account portal, open Manage API Access and follow OANDA's instructions to generate or manage a personal access token. Copy the account ID from the account you intend to use.

OANDA's authentication documentation describes token access and revocation. A token can authorize access to multiple accounts; verify access to each account you plan to use.

Store local configuration

Install the optional loader if needed: uv add python-dotenv. Create .env in the application directory and add it to .gitignore:

FIVETWENTY_OANDA_TOKEN=your-practice-token
FIVETWENTY_OANDA_ACCOUNT=your-account-id
FIVETWENTY_OANDA_ENVIRONMENT=practice

Replace the placeholders with your practice credentials. Keep real tokens out of source code, issues and shared notebooks. FiveTwenty reads environment variables; the load_dotenv() call below loads your .env file into that environment.

Verify account access

Save this as verify_access.py in the same directory as .env:

import asyncio

from dotenv import load_dotenv
from fivetwenty import AsyncClient, Environment

load_dotenv()


async def main() -> None:
    async with AsyncClient() as client:
        if client.config.environment != Environment.PRACTICE:
            message = "This tutorial requires a practice account"
            raise ValueError(message)
        response = await client.accounts.get_account_summary(client.account_id)
        account = response["account"]
        print(client.config.summary())
        print(f"Account currency: {account.currency}")
        print(f"Last transaction: {response['lastTransactionID']}")


if __name__ == "__main__":
    asyncio.run(main())

Run uv run verify_access.py, or python verify_access.py in an activated virtual environment. A successful run prints a configuration summary, the account currency and the last transaction ID. That confirms the token can read this account's summary.

The practice check uses the client's resolved configuration. By default, load_dotenv() keeps environment variables that are already set. If the client uses different values from your .env file, check your shell or process settings.

Other configuration sources

You can also pass credentials directly to the client constructor or use AccountConfig for a named, reusable configuration. The configuration guide explains which source takes priority and how to use custom environment-variable prefixes and configuration aliases.

Tokens use Pydantic's SecretStr to mask them when you display configuration, and SDK request logs redact Authorization headers. Keep the .env file private and avoid extracting or printing the token in your own logs.

If the check fails

• If client creation fails, check for missing or malformed configuration values.
• For a 401 or 403 response, check the token, resolved environment and account access.
• For a transport error, check the network, proxy, TLS settings and request timeout.

The authentication troubleshooting guide walks through these checks without exposing credentials in logs.

Next, create and close a practice trade.