# Kalshi historical data

Read the trades behind a settled market.

Get dated Kalshi trades for a market ticker. Our worked example reads an October Fed-rate contract, with contract counts, yes/no prices and a cursor for the next page.

Ten records from one settled contract, across two pages. The provider still reported more history after page two.

Provider in the worked example: Predexon via BlockRun.

Price shown for this route: 0.85¢ per request.

## Choose the history you need.

Trade records, candles and a complete exchange archive answer different questions.

Access and tested prices checked on 11 November 2026.

| Option | Price and billing | Strength | Good to know | Access |
| --- | --- | --- | --- | --- |
| Predexon via BlockRun (Through TaskFuel) | $0.0085 / request. Per request from your TaskFuel balance. Five rows requested per page in this test. | One TaskFuel balance; two trade-history pages inspected. | Full archive coverage and candlestick output were not tested. | TaskFuel |
| Predexon direct (Own API key) | Free trade endpoint. Documented as free and unlimited on every plan. Direct API authentication required. | A direct route for repeated trade-history queries. | Separate credentials; this test used the TaskFuel route. | Direct |
| Kalshi historical API (Direct exchange API) | See exchange documentation. Direct access under Kalshi’s API terms. The archive publishes cutoffs by data type. | Exchange-native historical trades and market candles. | Live and historical routes have different data windows; no direct historical call was tested. | Direct |

- [Predexon via BlockRun: provider reference](https://blockrun.ai/docs/api-reference/prediction-markets). Checked 2026-11-11.
- [Predexon direct: provider reference](https://docs.predexon.com/api-reference/kalshi/trades). Checked 2026-11-11.
- [Kalshi historical API: provider reference](https://docs.kalshi.com/getting_started/historical_data). Checked 2026-11-11.


## What it can do

### Read trades within a date range

A market ticker and Unix timestamps select the history. Each row includes trade ID, contract count, yes/no prices, taker side and trade time.

Price: $0.0085 / request.

### Continue with the returned cursor

The next page uses the same filters and the returned pagination key. The documented maximum is 500 rows per page; our test used five.

Price: $0.0085 / next page.


## A real example

Tested on 2026-11-11 through Predexon via BlockRun.

The first page returned five trades for the requested ticker. A second request returned five more with distinct trade IDs. Every timestamp fell within the requested window.

Your agent asked: Read five historical trades for the October Fed-rate contract above 4.25%

Returned in this test:

- Contract: Fed upper bound above 4.25% after the 28 October 2026 meeting
- First returned trade: 94 contracts; yes $0.01, no $0.99; taker side yes
- Trade time: 28 October 2026 at 17:49:34 UTC
- Pagination test: 5 + 5 records; 10 distinct trade IDs; more history available
- Provider contract: [Predexon Kalshi trades documentation](https://docs.predexon.com/api-reference/kalshi/trades)

Charged in this test: $0.0085.

The receipt is for page one. Page two cost another $0.0085. Two pages are a sample of one contract’s history.




## Try the API request

Use your TaskFuel API key as TASKFUEL_API_KEY in your terminal or server. Keep it out of browser code.

### 1. Check the price

This request checks the price without paying. Exact first-page request. start_time=1790812800 and end_time=1793232000 mean 1 October and 29 October 2026 at 00:00 UTC. Both parameters use seconds.

```sh
curl --fail-with-body https://app.taskfuel.ai/v1/call \
  -H "Authorization: Bearer $TASKFUEL_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-binary @- <<'TASKFUEL_REQUEST'
{
  "url": "https://blockrun.ai/api/v1/pm/kalshi/trades?ticker=KXFED-26OCT-T4.25&start_time=1790812800&end_time=1793232000&limit=5",
  "method": "GET",
  "dryRun": true
}
TASKFUEL_REQUEST
```

### 2. Run the request

Review amount_usd in the quote. To pay and run, change dryRun to false and add maxAmountUsd with that approved amount inside the same JSON object. Keep the inputs unchanged; this limit applies to one call.

### 3. Read the result

Read trades[].trade_id, ticker, count, yes_price, no_price, taker_side and created_time. Here count is contracts and 0.01 is $0.01 per contract, or 1¢. created_time is Unix seconds; created_time_ms is milliseconds. For the next page, URL-encode pagination.pagination_key and send it as pagination_key with the same filters. Each page is a new paid request. pagination.has_more was true on both tested pages.

[Read the Kalshi trade response contract](https://docs.predexon.com/api-reference/kalshi/trades)

[Setup and API reference](https://taskfuel.ai/docs/) · [Job polling and timeout recovery](https://app.taskfuel.ai/using.md)


## Connect your agent

Read the [setup guide](https://taskfuel.ai/docs/setup/) and choose the connection your agent supports.

Use one TaskFuel account and prepaid balance. Inspect the provider and request quote before a paid call, then review the returned response and recorded charge.



## When to use it

### Read prices before settlement.

Your agent retrieves dated trades for the chosen ticker and puts contract counts next to prices.

### Turn returned trades into rows.

An agent can write the returned JSON to CSV with trade ID, ticker, UTC time, count and yes/no prices. The tested result is a paginated JSON response.

### Separate trades from candles.

Individual fills describe executions. A candlestick series aggregates a time interval, while an order book describes available bids and asks. This example reads trades.




## Questions

### How do I get Kalshi historical data?

Select a market ticker and a UTC date range, then read trade pages. TaskFuel provides the tested Predexon route through BlockRun. Kalshi also documents its own historical API.

### Does this include every historical Kalshi market?

The test covers one settled October 2026 Fed-rate contract. It does not establish earliest coverage, complete market coverage or a full archive export.

### Why might older data be missing from Kalshi’s live API?

Kalshi separates live and historical data. Its historical cutoff endpoint publishes a separate cutoff for each data type. Older trades belong on the historical trades route; market candles have their own historical route.

### Can I get Kalshi historical data as CSV?

The tested endpoint returns JSON. Your agent can convert returned trade rows into CSV. A complete export requires paging through the selected range until the provider reports no more results.

### Can I use Predexon directly?

Yes. Predexon’s direct trade endpoint requires its own API key. Its documentation lists this endpoint as free and unlimited on every plan, so direct access may be a better fit for repeated exports.

### What did the historical-data requests cost?

Each five-row request cost $0.0085 on 11 November 2026. Two tested pages cost $0.017 total. Your agent shows the price for your own request before it pays.

## Related tool guides

- [Connect Kalshi research through MCP](https://taskfuel.ai/tools/kalshi-mcp/)
- [Read Polymarket historical data](https://taskfuel.ai/tools/polymarket-historical-data/)
- [Search current prediction market odds](https://taskfuel.ai/tools/prediction-market-api/)


## Reference

- [Read this guide on the website](https://taskfuel.ai/tools/kalshi-historical-data/).
- [Browse providers and capabilities](https://taskfuel.ai/catalog/).
- [Pricing](https://taskfuel.ai/pricing/).
- [Agent instructions](https://taskfuel.ai/llms.txt).
