Automated Bitcoin savings plan and withdrawal using Kraken, designed for Google Cloud Run.

This project contains a Node.js application that provides two main functionalities, packaged for deployment on Google Cloud Run:

• Buy Bitcoin: An endpoint (/buy) that buys a predefined amount of Bitcoin (default: 5 EUR) on Kraken. It fetches the current price, calculates the volume, checks your EUR balance, and places a limit order.
• Withdraw Bitcoin: An endpoint (/withdraw) that withdraws Bitcoin from your Kraken account to a pre-configured withdrawal address if the balance exceeds a certain threshold (default: 0.002 BTC).

Both actions send notifications to a Slack webhook. The application is designed to be triggered via HTTP POST requests to these endpoints and can be scheduled using Google Cloud Scheduler.

• Node.js: (v18+ recommended, as used in Dockerfile). Download from nodejs.org.
• npm: (Node Package Manager, comes with Node.js).
• Docker: Install Docker Desktop or Docker Engine. Follow the official Docker installation guide.
• Google Cloud SDK (gcloud CLI): Install and initialize it. Follow the official Google Cloud SDK installation guide. Ensure it's configured for your project.
• Google Cloud Project: You'll need a Google Cloud Project with billing enabled. Note the Project ID.
  • Enable the Cloud Run API and Artifact Registry API (or Google Container Registry API).
• Kraken Account: With API keys (key and secret) generated. Ensure the API key has permissions for:
  • Querying account balance
  • Querying ticker information
  • Creating buy/sell orders
  • Querying withdrawal information
  • Initiating withdrawals (for /withdraw endpoint)
• Kraken Withdrawal Address Key: For the /withdraw endpoint, you need to have a withdrawal address pre-configured in your Kraken account and know its 'key' or 'name'.
• Slack Webhook URL: For receiving notifications.

1. Clone the repository:

   git clone https://github.com/your-username/kraken-helper.git # Replace with your repo URL if forked
   cd kraken-helper

2. Install dependencies (for local development/testing):

   npm install

   This will install express, axios, dotenv, and kraken-api as defined in package.json. Dependencies for deployment are handled within the Docker build.

3. (Optional) Create a .env file for local development: Copy .env.example (if provided) or create a new .env file in the root directory with your credentials for local testing:

   KRAKEN_API_KEY="YOUR_KRAKEN_API_KEY"
   KRAKEN_API_SECRET="YOUR_KRAKEN_API_SECRET"
   SLACK_WEBHOOK_URL="YOUR_SLACK_WEBHOOK_URL"
   KRAKEN_WALLET_NAME="YOUR_KRAKEN_WALLET_KEY_NAME" # Only if testing withdraw
   # PORT=8080 # Optional for local development, defaults to 8080. Cloud Run provides its own PORT variable.

   Important: Do not commit your .env file to Git. It's included in .gitignore.

The application is configured using environment variables. Do not hardcode your credentials.

When deploying to Google Cloud Run, you will set these variables in the service's configuration.

Required for both endpoints:

• KRAKEN_API_KEY: Your Kraken API Key.
• KRAKEN_API_SECRET: Your Kraken API Secret.
• SLACK_WEBHOOK_URL: Your Slack incoming webhook URL.

Required only for the /withdraw endpoint:

• KRAKEN_WALLET_NAME: The 'key' or 'name' of your pre-configured withdrawal address in your Kraken account settings.

Security Note: For enhanced security, consider using Google Secret Manager to store your sensitive credentials and grant your Cloud Run service's service account access to these secrets.

Deployment involves building a Docker container image, pushing it to a container registry (like Google Artifact Registry or Google Container Registry), and then deploying it as a Cloud Run service.

Replace YOUR_PROJECT_ID, YOUR_REGION (e.g., us-central1), SERVICE_NAME (e.g., kraken-helper), and IMAGE_NAME with your actual details.

From the root of the kraken-helper directory. If you are on an Apple Silicon (M-series) Mac or another non-amd64 architecture, specify the platform:

docker build --platform linux/amd64 -t gcr.io/YOUR_PROJECT_ID/IMAGE_NAME:latest .
# Or for Artifact Registry:
# docker build --platform linux/amd64 -t YOUR_REGION-docker.pkg.dev/YOUR_PROJECT_ID/YOUR_REPO_NAME/IMAGE_NAME:latest .

Example: docker build --platform linux/amd64 -t gcr.io/my-gcp-project/kraken-app:latest .

Using Google Container Registry (GCR):

gcloud auth configure-docker
docker push gcr.io/YOUR_PROJECT_ID/IMAGE_NAME:latest

Using Google Artifact Registry: (Ensure you have an Artifact Registry Docker repository created)

gcloud auth configure-docker YOUR_REGION-docker.pkg.dev
docker push YOUR_REGION-docker.pkg.dev/YOUR_PROJECT_ID/YOUR_REPO_NAME/IMAGE_NAME:latest

Deploy the image to Cloud Run, ensuring it requires authentication. Cloud Run automatically sets the PORT environment variable; do not set it manually here.

gcloud run deploy SERVICE_NAME \
  --image gcr.io/YOUR_PROJECT_ID/IMAGE_NAME:latest \
  # Or for Artifact Registry: --image YOUR_REGION-docker.pkg.dev/YOUR_PROJECT_ID/YOUR_REPO_NAME/IMAGE_NAME:latest \
  --platform managed \
  --region YOUR_REGION \
  --set-env-vars KRAKEN_API_KEY="YOUR_KRAKEN_API_KEY",KRAKEN_API_SECRET="YOUR_KRAKEN_API_SECRET",SLACK_WEBHOOK_URL="YOUR_SLACK_WEBHOOK_URL",KRAKEN_WALLET_NAME="YOUR_KRAKEN_WALLET_KEY_NAME" \
  --no-allow-unauthenticated

• --no-allow-unauthenticated: This is crucial. It ensures your Cloud Run service is private and requires IAM authentication for invocation. This is recommended for secure, automated tasks.
• After deployment, Cloud Run will provide a service URL. This URL will be private.

To automate the execution of the buy and withdraw actions, use Google Cloud Scheduler to send authenticated HTTP POST requests to your private Cloud Run service endpoints.

This service account will be used by Cloud Scheduler to invoke your Cloud Run service.

gcloud iam service-accounts create SCHEDULER_INVOKER_NAME 
  --display-name="Service Account for Cloud Scheduler Invocation"

Replace SCHEDULER_INVOKER_NAME (e.g., kraken-scheduler-invoker). Note the full email of the created service account (e.g., SCHEDULER_INVOKER_NAME@YOUR_PROJECT_ID.iam.gserviceaccount.com).

Allow the new service account to invoke your Cloud Run service.

gcloud run services add-iam-policy-binding SERVICE_NAME \
  --member="serviceAccount:SCHEDULER_INVOKER_EMAIL" \
  --role="roles/run.invoker" \
  --region=YOUR_REGION \
  --platform=managed

Replace SERVICE_NAME, SCHEDULER_INVOKER_EMAIL, and YOUR_REGION.

Create a job for each endpoint (/buy and /withdraw).

Example for /buy job (e.g., every 2 days at 08:00 UTC):

gcloud scheduler jobs create http YOUR_BUY_JOB_NAME \
  --schedule="0 8 */2 * *" \
  --uri="YOUR_CLOUD_RUN_SERVICE_URL/buy" \
  --http-method=POST \
  --message-body="{}" \
  --oidc-service-account-email="SCHEDULER_INVOKER_EMAIL" \
  --location=YOUR_SCHEDULER_REGION \
  --description="Trigger Kraken Bitcoin buy operation"

Example for /withdraw job (e.g., 1st of month at 10:00 UTC):

gcloud scheduler jobs create http YOUR_WITHDRAW_JOB_NAME \
  --schedule="0 10 1 * *" \
  --uri="YOUR_CLOUD_RUN_SERVICE_URL/withdraw" \
  --http-method=POST \
  --message-body="{}" \
  --oidc-service-account-email="SCHEDULER_INVOKER_EMAIL" \
  --location=YOUR_SCHEDULER_REGION \
  --description="Trigger Kraken Bitcoin withdrawal check"

• Replace placeholders like YOUR_BUY_JOB_NAME, YOUR_CLOUD_RUN_SERVICE_URL, SCHEDULER_INVOKER_EMAIL, and YOUR_SCHEDULER_REGION.
• The --message-body="{}" sends an empty JSON object, which might be required by some setups.
• The --oidc-service-account-email ensures the job authenticates correctly to your private Cloud Run service.
• Schedules are in UTC by default. Use --time-zone to specify a different one if needed.
• You can use crontab.guru to help create your cron expressions.

1. Ensure you have a .env file with your credentials.
2. Run the server locally:

   npm start

   The server will start, typically on http://localhost:8080.
3. You can then use tools like curl or Postman to send POST requests to your local endpoints:
   • curl -X POST http://localhost:8080/buy
   • curl -X POST http://localhost:8080/withdraw

Check the console output for logs and Slack for notifications.