Kalshi

Kalshi is a regulated prediction market exchange. ingestr supports Kalshi as a public read-only source for exchange status, series, events, markets, order books, trades, candlesticks, and historical market data.

No API key is required for the supported tables.

URI format

kalshi://

Optional URI parameters are passed to supported API filters. Examples:

• kalshi://?status=open
• kalshi://?series_ticker=KXHIGHNY
• kalshi://?event_ticker=<event-ticker>
• kalshi://?ticker=<market-ticker>
• kalshi://?tickers=<ticker-1>,<ticker-2>

How it works

ingestr reads Kalshi's public Trade API at:

https://external-api.kalshi.com/trade-api/v2

Each --source-table maps to one public endpoint. Rows include selected stable columns plus a raw JSON column containing the full source payload. Numeric fixed-point fields from Kalshi are preserved as strings unless the API returns a numeric JSON value.

Interval behavior

Quick summary: intervals narrow the API request for market discovery, trades, and candlesticks. For markets, the interval means "markets created in this date range", not "markets updated in this date range". For trades, it means "trades in this time range". For candlesticks, it means "candlestick periods in this time range".

Important: using an incremental strategy with intervals on markets can miss markets that were created outside the interval but updated inside it. Kalshi does expose min_updated_ts, so ingestr can ask for markets updated after a start time. However, Kalshi does not expose a matching max_updated_ts, ignores undocumented max_updated_ts, and rejects min_updated_ts when combined with most other filters such as status=open. A reliable "updated between start and end" sync would need min_updated_ts plus client-side filtering for the interval end.

When an endpoint supports server-side time filtering, ingestr pushes intervals into the API call:

Table	API interval params
markets	min_created_ts, max_created_ts as Unix seconds. These filter market creation time, not updated_time.
market_trades	min_ts, max_ts as Unix seconds. These filter trade creation/execution time.
historical_trades	min_ts, max_ts as Unix seconds. These filter trade creation/execution time.
market_candlesticks	required start_ts, end_ts as Unix seconds. These filter candlestick periods.
market_candlesticks_batch	required start_ts, end_ts as Unix seconds. These filter candlestick periods.

market_candlesticks and market_candlesticks_batch require both --interval-start and --interval-end because Kalshi requires those parameters.

Example

ingestr ingest \
  --source-uri 'kalshi://?status=open' \
  --source-table markets \
  --dest-uri 'duckdb:///kalshi.duckdb' \
  --dest-table kalshi.markets

Use a ticker and event_ticker returned by markets for single-market tables:

ingestr ingest \
  --source-uri 'kalshi://?ticker=<market-ticker>' \
  --source-table market_orderbook \
  --dest-uri 'duckdb:///kalshi.duckdb' \
  --dest-table kalshi.market_orderbook

Candlestick example:

ingestr ingest \
  --source-uri 'kalshi://?series_ticker=<series-ticker>&ticker=<market-ticker>&period_interval=60' \
  --source-table market_candlesticks \
  --interval-start '2026-01-01' \
  --interval-end '2026-01-02' \
  --dest-uri 'duckdb:///kalshi.duckdb' \
  --dest-table kalshi.market_candlesticks

Tables

Table	Required URI params	Optional URI params	PK	Inc Key	Details
exchange_status	-	-	-	-	Exchange active/trading active flags and estimated resume time.
exchange_schedule	-	-	-	-	Exchange schedule payload in raw.
exchange_announcements	-	-	id	created_time	Public exchange announcements.
series	-	category, tags	ticker	updated_time	Series metadata.
series_by_ticker	series_ticker	-	ticker	-	One series by ticker.
events	-	series_ticker, status, with_nested_markets	event_ticker	updated_time	Events and optional nested markets.
event_by_ticker	event_ticker	-	event_ticker	-	One event by ticker.
markets	-	event_ticker, series_ticker, status, tickers, mve_filter, min_updated_ts, max_close_ts, min_close_ts, min_settled_ts, max_settled_ts	ticker	updated_time	Market discovery with prices, volume, open interest, status, and raw.
market_by_ticker	ticker	-	ticker	-	One market by ticker.
market_orderbook	ticker	-	-	-	Current YES/NO bid ladders for one market.
market_orderbooks	tickers	-	-	-	Batch order books for comma-separated tickers.
market_trades	-	ticker, is_block_trade	trade_id	created_time	Public trades. Supports interval pushdown.
market_candlesticks	series_ticker, ticker, period_interval	include_latest_before_start	end_period_ts	end_period_ts	Candlesticks for one market. Requires intervals.
market_candlesticks_batch	market_tickers, period_interval	include_latest_before_start	-	-	Batch candlesticks for up to 100 market tickers. Requires intervals.
historical_markets	-	tickers, event_ticker, series_ticker, status	ticker	-	Archived historical markets.
historical_trades	-	ticker, is_block_trade	trade_id	created_time	Historical trades. Supports interval pushdown.

Notes

• Use status=open to find currently populated live markets. Some series return zero events or markets, so a reliable smoke-test flow is: ingest markets with status=open, take a returned ticker and event_ticker, derive the series ticker from the event ticker prefix, then call market_by_ticker, event_by_ticker, order book, trade, and candlestick tables.
• market_candlesticks and market_candlesticks_batch require both --interval-start and --interval-end. Use a short window around the selected market's active period to avoid empty results.
• Batch order books use the tickers URI parameter, while batch candlesticks use market_tickers.
• Kalshi order books return YES and NO bids; asks are implied by binary market mechanics.
• When loading into DuckDB, --schema-naming direct is currently the safest option for these tables because many provider field names are mixed case or already provider-specific.
• Authenticated trading, portfolio, order, account, and RFQ endpoints are not supported.