cryptofeed
Websocket feed handler that normalises trades and books across exchanges
🧩 Build it Free 01 in Data & feeds 🆕 NEW TODAY
2,899 0 copies · 7 days AGPL-3.0updated 8 Sep '26
Overview
Cryptofeed is a Python websocket feed handler that normalises trades, order books, tickers, and funding data across many cryptocurrency exchanges.
- Fordevelopers who want one normalized data shape across many crypto exchanges instead of writing a parser per venue.
- NeedsPython 3.12+; no key for public channels, an exchange key only for authenticated order/balance streams.
- Runsan asyncio Python library (pip install cryptofeed) with a FeedHandler you add exchange subscriptions to, plus optional storage backends (Redis, Kafka, PostgreSQL, and more).
- Limitevery callback must be a coroutine — a plain synchronous function has to be wrapped, or the handler drops it.
Vibe it
Paste this into your AI agent. Write the first line, or take a suggestion; the copy carries your words.
Source: https://github.com/bmoscon/cryptofeed — clone it and read its README and examples before writing anything; its documented setup beats what you remember. Stack: an asyncio Python library — a FeedHandler with one subscription per exchange and channel, delivering normalised objects to async callbacks. Needs: Python 3.12+; no key for public trade/book/ticker channels, an exchange API key only for authenticated streams. Start: `pip install cryptofeed`, then add one feed (e.g. Coinbase) with a TRADES callback and call fh.run() before adding more exchanges. Limit: every callback must be a coroutine — wrap a synchronous one in cryptofeed.callback.ExecutorCallback or the handler drops it. Done means: you show me normalised trade prints from one exchange running in my terminal, and the feed code you wrote. Before you start, ask me: which exchange and symbol I want to start with, and whether I need a storage backend (Redis, Kafka, PostgreSQL) or just console output.
