Dynamic Cryptocurrency Pricing

This project showcases a lightweight dynamic pricing engine for merchants that sell goods or services while getting paid in crypto. It loads crypto market data, derives risk and momentum signals, and outputs adaptive price recommendations for a configurable product catalog.

Features

• Pull historical candles from CoinMarketCap (API key) or fall back to the bundled CSV sample.
• Compute rolling volatility, momentum, moving averages, and spread risk indexes.
• Combine business rules (target margin, elasticity curve, volatility guardrails) into a final suggested markup.
• CLI to simulate the pricing engine on top of sample data or live API calls.
• Modular architecture (config parsing, data sources, signal generation, pricing strategies) ready for extension.

Quick Start

Streamlit dashboard (live data)

uv venv
source .venv/bin/activate
cp .env.example .env
$EDITOR .env   # fill in COINMARKETCAP_API_KEY
make install-ui
make dashboard

CLI simulation workflow

uv run python -m dynamic_pricing.cli \
  --config examples/configs/example_config.yaml \
  --data examples/data/sample_market_data.csv \
  --market-condition bull \
  --verbose

uv run python -m dynamic_pricing.cli \
  --config examples/configs/example_config.yaml \
  --market-condition balanced \
  --verbose

CoinMarketCap API Key

The live data source expects an API key. Provide it via the COINMARKETCAP_API_KEY environment variable (recommended) or add api_key under data_source in your YAML config. The key is only used to resolve asset and fiat metadata; price candles are fetched from CoinMarketCap's historical endpoint.

The CLI automatically loads environment variables from a local .env file (or the path defined in DYNAMIC_PRICING_ENV_FILE). For convenience, copy .env.example to .env and fill in your actual API key.

Repository Layout

.
├── examples/
│   ├── configs/             # YAML configs describing catalog, guardrails and API options
│   └── data/                # Sample market data for offline testing
├── src/dynamic_pricing/     # Pricing engine source code
├── tests/                   # Pytest suite covering the core engine logic
└── README.md                # This file

Market Conditions

Each strategy tweaks the markup curve. Swap --market-condition to experiment:

uv run python -m dynamic_pricing.cli \
  --config examples/configs/example_config.yaml \
  --market-condition <strategy>

• balanced – original volatility-aware blend of momentum and guardrails (default):

  uv run python -m dynamic_pricing.cli --config examples/configs/example_config.yaml --market-condition balanced

• bull – pushes markups higher when upward momentum is present:

  uv run python -m dynamic_pricing.cli --config examples/configs/example_config.yaml --market-condition bull

• bear – prioritizes downside protection and margin preservation:

  uv run python -m dynamic_pricing.cli --config examples/configs/example_config.yaml --market-condition bear

• lateral – compresses markups when price action chops sideways:

  uv run python -m dynamic_pricing.cli --config examples/configs/example_config.yaml --market-condition lateral

• penetration – intentionally lowers markups to grab market share:

  uv run python -m dynamic_pricing.cli --config examples/configs/example_config.yaml --market-condition penetration

• competitor – aligns with the configured competitor reference price:

  uv run python -m dynamic_pricing.cli --config examples/configs/example_config.yaml --market-condition competitor

Configuration Notes

• Product catalog – define name, desired target_margin, price elasticity, and optionally competitor_name (e.g., Binance, Kraken). The computed markup is applied to the latest market price pulled from your data source, and the competitor strategy fetches the rival quote from a stubbed data source by name.
• Competitor provider – pass --competitor-provider coinmarketcap (or set DYNAMIC_PRICING_COMPETITOR_PROVIDER=coinmarketcap) plus a COINMARKETCAP_API_KEY to source quotes directly from CoinMarketCap's market-pairs endpoint. Leave it as stub to use the bundled deterministic map.
• Guardrails – min_markup/max_markup keep recommendations inside a safe band, while volatility_floor/volatility_ceiling normalize risk signals.
• Data source – choose coinmarketcap or csv, pick the asset (symbol or CoinMarketCap ID), vs_currency, lookback_hours, and optionally api_key/api_url.
• Smoothing window – controls the rolling lookback (in hours) used for momentum, volatility, and averages.
• Streamlit dashboard – streamlit_app.py loads the default YAML config, pulls live candles from CoinMarketCap using your API key, and lets you tune asset, quote currency, lookback, smoothing, guardrails, competitor names, and strategy without editing files.

Extending

• Replace the CSV data source with another API by implementing BaseMarketDataSource.
• Tweak the VolatilityAwareStrategy or plug in a new strategy subclass.
• Integrate the engine into another service via the PriceEngine class.
• Prototype guardrails and business rules visually via streamlit_app.py, which exposes sparkline charts and sidebar controls for strategy, guardrails, and competitor settings.

Tests

uv run pytest

The suite focuses on the pricing engine behavior and guardrails.