Explorer Dashboard

ccxt-pandas ships an optional Streamlit dashboard that lets you browse any CCXT exchange method, see the resulting DataFrame, and copy the equivalent Python snippet — useful for prototyping before writing scripts.

A hosted version is at ccxt-explorer.com.

Install

pip install "ccxt-pandas[explorer]"

Pulls in streamlit, plotly, and Pillow.

Run

ccxt-pandas-explorer
# or
uv run ccxt-pandas-explorer

The CLI launches Streamlit on http://localhost:8501 against the bundled app at ccxt_pandas_explorer/app.py.

What you can do

  • Pick any exchange from the sidebar (~120 supported via CCXT).

  • Pick any fetch_* method (the dropdown is built by introspecting exchange.has so only methods that actually work on that exchange show up).

  • Pick symbols from the markets table (single- or multi-row selection depending on whether the method takes symbol or symbols).

  • Set parameters — timeframe, since, limit etc. are auto-detected from the method signature.

  • See the DataFrame post-conversion (UTC timestamps, numeric coercion, schema-validated shape).

  • Copy the equivalent code snippet to paste into your own project.

  • Plot the result with Plotly (line/bar/scatter, optional aggregation, facets, color, size).

Adding credentials

The sidebar shows fields for whichever auth params the chosen exchange requires (apiKey, secret, password, …). They are kept in Streamlit session state — not persisted to disk. For private endpoints, paste them in; for public-only endpoints (OHLCV, order books, trades), leave them blank.

Sandbox mode

If the exchange supports a sandbox/testnet, a toggle appears in the sidebar. Recommended for any test of order placement.

Source

The app source is one self-contained file — see ccxt_pandas_explorer/app.py. Fork it, extend it, run your own version.