Kalshi Weather Markets Scraper — every city, every bracket, $0.80/1,000 markets

Live bracket prices and recent settlements for every Kalshi weather market: daily high/low temperature for all listed cities (NYC, Chicago, Miami, Austin, Denver, LA, Philadelphia, Houston, Seattle, Toronto, London, and more), plus rain, snow and hurricane markets — 360+ series discovered automatically, not a hardcoded city list.

Built on Kalshi's current public trade API (the new *_dollars schema) — the API migration that quietly broke older scrapers. Plain HTTP, no login, no API key, no browser, no proxies. Runs finish in seconds.

$0.80 per 1,000 market records + $0.001 per start. maxMarkets is a hard cost cap.

Who uses this

• Weather traders — pull every open bracket with implied probability, volume and open interest before placing an order, instead of clicking through city pages.
• Model builders — join settledValue (the actual settled temperature) with your forecast data to backtest a strategy per city and station.
• Dashboards & alerts — schedule it and pipe fresh odds into Sheets, Slack or your own model.
• AI agents — stable flat schema, predictable cost, one-call market snapshot.

What you get per market

Field	Example	Notes
seriesTicker / seriesTitle	KXHIGHNY	which weather series
eventTicker	KXHIGHNY-26AUG29	one event = one city-day
marketTicker	KXHIGHNY-26AUG29-B85.5	one row per bracket
bracket	85° to 86°	human-readable strike
strikeType, floorStrike, capStrike	between, 85, 86	machine-readable strike
yesBid / yesAsk / noBid / noAsk / lastPrice	0.56	dollars, parsed to numbers
impliedProbability	0.565	bid/ask mid (falls back to last price)
volume, volume24h, openInterest, liquidity		activity metrics
status, result	finalized, no	settled markets carry the result
settledValue	77	the actual settled temperature
station	CLINYC	exact settlement weather station
openTime / closeTime / expectedExpirationTime	ISO	timing
marketUrl		one-click open on Kalshi

Every field is always present (nullable, never missing). marketTicker is a stable dedupe key.

Input

{
  "mode": "temperature",
  "cities": ["NY", "Chicago"],
  "status": "open",
  "maxMarkets": 500
}

• mode — temperature (every high/low/avg temp series) or all-weather (the entire Climate & Weather category: rain, snow, hurricanes too).
• cities — optional keyword filter (["NY"], ["Denver", "Miami"]). Empty = all cities.
• seriesTickers — exact series override, e.g. ["KXHIGHNY", "KXRAINDNYC"].
• status — open (live prices), settled (recent results + settled temps), or both.
• settledLookbackDays — how far back settled markets go (default 3).
• maxMarkets — hard cost cap ($0.80/1,000).

How to schedule this Actor (recommended)

Weather odds move by the minute — this is built to run on a schedule:

1. Open the Actor → Try for free, set your cities and status.
2. Save as a task → Schedules → e.g. */30 * * * * (every 30 min while markets are open) or 0 * * * * hourly.
3. Add an integration (Google Sheets, Slack, webhook) so fresh brackets land where your model reads them.
4. For settlement tracking, a second daily task with "status": "settled" gives you yesterday's settled temperatures per station.

Monitoring 10 cities every 30 minutes typically costs a few dollars a month — the run is one row per bracket, and maxMarkets caps every run.

Backtesting recipe

Run "status": "settled", "settledLookbackDays": 90, "seriesTickers": ["KXHIGHNY"] once: you get every bracket of every NYC high-temp market for the last 90 days with result and settledValue — join against your forecast history to measure edge per city.

For AI agents & MCP

• Minimal input that just works: {} — scrapes all open temperature markets.
• Predictable cost: exactly $0.0008 × markets + $0.001 per run.
• Stable schema, marketTicker dedupe key, RUN_SUMMARY in the key-value store.

Related Actors

• Brand Ads Spy — Google + LinkedIn Ads in One Feed — cross-network competitor ad intelligence.
• Superteam Earn Scraper — bounty and project boards with EV per entry.

Fair use & data source

Only public market data from Kalshi's public trade API (the same data shown on kalshi.com). Reasonable rate limits (~2 req/s). No account data, no personal data, no order placement — this is a read-only data tool, not a trading bot. Not investment advice.