YFinance Proxy Service

A lightweight FastAPI microservice that wraps yfinance, exposing RESTful endpoints for market data, perfect for non-Python projects, microservice architectures, and monitorable deployments.

Why Use This?

• Language-agnostic: Get Yahoo Finance data via HTTP, no python dependency.
• Production-ready: Includes Prometheus metrics and health checks.
• Simple setup: Run with Docker, Poetry, or Docker Compose (Grafana/Prometheus included).
• Extendable: Built on FastAPI; easy to add routes or middleware.

Features

Feature	Description
Quote API	Fetch latest market quotes (OHLCV) for ticker symbols.
Historical API	Retrieve historical data within a date range.
Info API	Get company fundamentals (sector, market cap, etc.).
Health Check	Simple /health endpoint to verify service status.
Prometheus Metrics	/metrics endpoint for request count, errors, latency.

API Endpoints

Endpoint	Description	Example
GET /quote/{symbol}	Latest quote for a symbol	/quote/AAPL
GET /historical/{symbol}?start=YYYY-MM-DD&end=YYYY-MM-DD	Historical OHLCV data	/historical/AAPL?start=2024-01-01&end=2024-02-01
GET /info/{symbol}	Company details	/info/TSLA
GET /health	Health check	/health
GET /metrics	Prometheus metrics	/metrics

Quick Start

Run with Docker

docker pull ghcr.io/vorckea/yfinance-service:latest
docker run -p 8000:8000 ghcr.io/vorckea/yfinance-service:latest

Then visit: http://localhost:8000/docs: Interactive Swagger UI

Local Development (Poetry)

poetry install
poetry run uvicorn app.main:app --reload

With Prometheus + Grafana

docker compose up --build

Access:

• API: http://localhost:8000
• Prometheus: http://localhost:9090
• Grafana → http://localhost:3000

Relevant files:

• Prometheus config: monitoring/infra/prometheus.yml
• Alert rules: monitoring/infra/alert.rules.yml
• Dashboards: monitoring/dashboards/

Usage Example

Get the latest quote for Apple:

curl http://localhost:8000/quote/AAPL

Get company info for Tesla:

curl http://localhost:8000/info/TSLA

Monitoring Example:

Prometheus query for average latency (5-minute window):

rate(request_latency_seconds_sum[5m]) / rate(request_latency_seconds_count[5m])

Contributing

Contributions are welcome!

• Open an issue for bugs or new features.
• Look for good first issue labels to get started.
• Please see CONTRIBUTING.md

License

MIT License. See LICENSE for details.

Author

Aksel Heggland Schrader (@Vorckea)