Kalshi Screener MCP
This read-only MCP server lets AI apps screen Kalshi markets, inspect market and event data, review historical settlement evidence, and manage screens, watchlists, and alerts.
Search and filter open markets by price, bid/ask, spread, close time, listing age, volume, open interest, daily change, event favorite/size, and event YES totals.
Get detailed data for a market, including prices, order book top, rules, settlement sources, related event markets, and a research prompt.
Compare all open markets in an event ranked by YES price with YES bid/ask totals.
Retrieve historical evidence for a market from similar past events in the same series, including sample sizes and confidence ranges.
Save, delete, list, and run named screens, including presets shared by the Chrome extension.
Get live prices for the extension watchlist and list recent condition alerts from the extension monitor.
Allows deploying the MCP server as a self-hosted remote endpoint on Cloudflare Workers (with Wrangler-managed storage and an access-key secret), so web-based AI apps can connect to it over a URL while streaming live Kalshi market screening results.
Exposes the screener as a remote MCP server that can be used from Meta's Muse chat assistant by providing the server URL and access key, enabling market screening and saved-screen/alerts queries from the chat.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Kalshi Screener MCP Find markets priced 80c or higher closing in the next 48 hours"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Kalshi Screener: Chrome extension + MCP server
For research and education. Kalshi Screener is a free, open-source, read-only tool. It never places trades, gives no buy or sell advice and is not financial advice. It is an independent project, not affiliated with, endorsed by or sponsored by Kalshi. Use it at your own risk.
Screen and monitor Kalshi prediction markets with your own conditions, get alerts when a market matches, and ask your AI about them.
Scanner | Alerts | Bring your own AI |
|
|
|
This repository has two parts. Use either one or both.
Chrome extension (
extension/): a side panel that screens public market data, repeats your screen in the background and alerts you. No Kalshi account or API key needed.MCP server (everything else): lets Claude, ChatGPT, Cursor, VS Code and other AI apps search markets and read your saved conditions, watchlist and alerts.
No ads, no analytics, no hosted service. Everything runs on your computer (or your own Cloudflare account).
Install the Chrome extension
The extension is not on the Chrome Web Store. Install it from this repository in developer mode:
Download kalshi-screener-extension-0.3.0.zip from Releases (or download this repository and use its
extensionfolder).Unzip it to a folder you will keep (Chrome loads the extension from that folder, so do not delete it).
Open
chrome://extensionsin Chrome.Turn on Developer mode (top right).
Click Load unpacked and choose the unzipped folder (the one that contains
manifest.json).Pin Kalshi Screener from the puzzle icon and click it to open the side panel.
To update, download the new zip, replace the folder contents and click the reload icon on the extension card in chrome://extensions. Your settings stay in Chrome.
Chrome may show a notice about developer mode extensions. That is normal for extensions installed this way. You can read every line of the code in extension/.
Related MCP server: mcp-kalshi
Install the MCP server
Works with Claude (Desktop, Code and web), ChatGPT, Grok, Cursor and VS Code. Paired with the Chrome extension, your saved conditions, watchlist and alerts are shared with the AI, and every result has an Open in Kalshi Screener link.
Read-only. It uses Kalshi's public market data, never places or cancels orders, and gives no buy or sell advice.
Ask things like:
"Find markets priced 80c or higher that close in the next 48 hours."
"Which markets were listed in the last day?"
"Show every market in this event ranked by price."
"Run my 'Closing soon' screen and list today's alerts."
Desktop AI apps (runs on your computer)
Requires Node.js 20 or later. The server is not published to npm; it runs from this folder.
Download this repository (on GitHub: Code, Download ZIP) and unzip it.
Open a terminal in the unzipped folder and run
npm installonce.Point your AI app at
bin/cli.jsin that folder. ReplaceC:/path/to/kalshi-screener-mcpbelow with your folder's full path (forward slashes work on Windows too).
Claude Desktop, Cursor, VS Code and other MCP apps. Add this to the app's MCP settings (in Claude Desktop: Settings, Developer, Edit Config):
{
"mcpServers": {
"kalshi-screener": { "command": "node", "args": ["C:/path/to/kalshi-screener-mcp/bin/cli.js"] }
}
}Claude Code:
claude mcp add kalshi-screener -- node "C:/path/to/kalshi-screener-mcp/bin/cli.js"Restart the app. In the Chrome extension, Settings, AI Agents (MCP) builds the same settings from the folder path you paste and has a copy button.
Share data from the Chrome extension
In the extension, open Settings, AI Agents (MCP), turn on Share with the MCP server on this computer and press Check and share now. The extension then sends your condition presets, watchlist and recent alerts to 127.0.0.1 whenever they change and after each background check. API keys and AI settings are never sent. Only the extension can write to this local endpoint; web pages are refused.
The first extension that shares data is paired with the server, so other installed extensions cannot overwrite your data. If you reinstall the extension and it gets a new ID, start the server once with --pair (for example node bin/cli.js --pair in the server folder, or add "--pair" to args and remove it afterwards).
Open in Kalshi Screener links use chrome-extension://. Some desktop apps cannot open them directly; copy the link into Chrome's address bar instead.
Web AI apps (ChatGPT, Grok, Meta Muse, Claude on the web)
These apps connect to a web address, so deploy your own free copy to Cloudflare Workers.
Create a free Cloudflare account.
Deploy this folder. In the unzipped folder, run:
npm install npx wrangler deployWrangler creates the storage it needs on first deploy.
Choose a long random access key and save it:
npx wrangler secret put ACCESS_KEYAdd the server to your AI app as a custom connector or MCP server:
https://kalshi-screener-mcp.<your-subdomain>.workers.dev/mcp?key=<your access key>ChatGPT: Plugins, Add custom MCP server, paste the URL.
Claude on the web: Settings, Connectors, Add custom connector, paste the URL.
Meta Muse: tell Muse the server URL in chat and store the access key when it asks.
Grok (xAI API): pass the URL as a remote MCP tool, or send the key as
Authorization: Bearer <key>.
Optional: set EXTENSION_ID (npx wrangler secret put EXTENSION_ID) to your extension ID so remote results include Open in Kalshi Screener links.
Cloudflare's free plan allows about 50 outbound requests per call. Very broad searches over every open market can exceed that; the server then asks you to narrow the search (for example scope: "closing_24h"). The local server has no such limit.
How Your AI Combines Conditions from Plain-Language Prompts
You do not have to rely only on pre-built presets. When you ask a question in plain language, your AI automatically combines the filter parameters in search_markets, and can even save new screens for you:
Ad-hoc Multi-Condition Screening (AI combines filters automatically): Ask your AI in any language:
"Find Kalshi markets closing within 24 hours with a YES price of 80c or higher, a bid-ask spread under 5%, and at least 500 contracts of 24h volume."
"Show only the #1 favorite market in each event where the event has at least 4 outcomes and closes within 7 days."
"Find multi-outcome events where the sum of all YES ask prices is below 98c." Your AI translates your prompt into
search_marketsfilter parameters (scope,text,min_yes_price,max_yes_price,min_yes_bid,max_yes_ask,max_hours_to_close,min_hours_to_close,listed_within_hours,min_volume_24h,min_open_interest,max_spread_pct,min_change_pct,max_change_pct,event_favorite_only,min_event_markets,max_event_yes_ask_total,sort) and runs the scan live.
Save Any Prompt-Built Screen by Voice/Text (
save_screen):Tell your AI: "Save that filter combination as 'Closing Soon Favorites'."
Your AI calls
save_screenso you can reuse it anytime.
Run Your Saved or Chrome Extension Presets (
list_saved_screens,run_saved_screen):Presets you saved in the Kalshi Screener Chrome extension (when MCP sharing is enabled) and screens saved via
save_screencan both be run on demand: "Run my 'Closing Soon Favorites' screen and summarize the top matches."Inside the Chrome extension itself, typing a natural-language search in the bottom Kalshi Screener Chat and pressing
SendorDraftalso builds full boolean condition stacks (A AND (B OR C) AND NOT D) across all 45 Scanner Lab templates.
Tools
Tool | What it does |
| Screen open markets by YES price, bid and ask, hours to close, listing age, 24h volume, open interest, spread, change versus the previous day, event favorite, event size and event YES totals. Sort by volume, closing soon, price or newest. |
| One market: prices, top of the order book, rules, settlement sources, its event and |
| All open markets in an event ranked by YES price, with YES bid and ask totals. |
| Historical evidence for one market from settled events in its own series. See below. |
| Save or remove named filter sets. |
| Screens saved here plus presets shared by the extension. |
| Run a saved screen or an extension preset by name. |
| Live prices for the extension watchlist. |
| Recent condition matches from the extension's background monitor. |
Prices are dollars from 0 to 1, the same as the implied probability of YES. Scopes: open_markets (default, excludes multivariate combo markets), closing_24h, closing_7d, multivariate_only.
Historical evidence
Ask your AI "Will this go up?" or "Is this likely?" about a market and it can call get_market_evidence. The tool looks at the last 20 settled events in the same Kalshi series (up to 40) and reports:
Similar price history: in past events, markets priced within 5c of this market's current price at the same time before close, and how many of them settled YES. One market per past event is counted, so related outcomes are not counted twice.
Favorite record: how often the highest priced market in each past event won.
Recent events: what won, what the favorite was, and the prices at the time.
Every rate comes with its sample size and a 95% range (Wilson score interval), plus a confidence label: very_low under 10 comparable events, low under 30, medium under 100. Prices are the mid of the YES bid and ask when the spread is 10c or less, otherwise the last trade. Older events come from Kalshi's /historical endpoints.
This is a summary of how a series settled in the past. It is not a prediction, and a small sample says little. Each event can take one Kalshi request, so on a free Cloudflare Worker a long history may stop early; the result says so in data_notes, and the local server has no such limit.
Add browser research (optional)
Most AI apps can search the web on their own. News sites, X and forums often need a login or block bots. To let a desktop AI app read those pages in your own signed-in Chrome, add Microsoft's free Playwright MCP next to this server:
Install the Playwright MCP Bridge Chrome extension.
Use these MCP settings (the extension's Research with AI window has a copy button):
{
"mcpServers": {
"kalshi-screener": { "command": "node", "args": ["C:/path/to/kalshi-screener-mcp/bin/cli.js"] },
"playwright": { "command": "npx", "args": ["@playwright/mcp@latest", "--extension"] }
}
}Ask, for example: "Find markets over 80c closing within 24 hours, then research the top one with its research_prompt."
LM Studio (0.3.17 or later) reads the same format from %USERPROFILE%/.lmstudio/mcp.json on Windows and ~/.lmstudio/mcp.json on macOS and Linux, so local models can use both servers. Other options: Claude in Chrome for Claude paid plans and ego lite for Claude Code, Codex and Cursor. These are third-party tools, not affiliated with Kalshi Screener. They can read any site you are signed in to, so only connect accounts you are comfortable letting the AI read. The research prompt tells the AI to read only and never post or trade.
Command line options
node bin/cli.js MCP over stdio (default)
node bin/cli.js --http MCP over HTTP at http://127.0.0.1:47811/mcp
--port <n> Port for the extension bridge and --http (default 47811)
--state <path> State file (default ~/.kalshi-screener/mcp-state.json)
--no-bridge Do not accept data from the Chrome extension
--pair Pair with the next extension that shares dataKALSHI_SCREENER_EXTENSION_ID sets the extension ID used in links before the extension has shared data.
Questions
For bugs and ideas, open an issue. Other questions: contact@axnli.com.
License
Open source under the MIT License. Free for personal and commercial use. See PRIVACY.md for what the extension and server do with your data.
Built by AXNli, a team that designs data and AI workflows.
Available Tools
10 toolsdelete_screenDelete a saved screenADestructiveIdempotent
Delete a screen saved through this server. Screens saved in the Chrome extension are not affected.
Inputs: name (required).
get_eventCompare markets in an eventBRead-only
All open markets in one Kalshi event ranked by YES price, with YES bid and ask totals.
Inputs: event_ticker (required).
get_marketGet a Kalshi marketARead-only
Prices, top of the order book, rules, settlement sources and the other markets in the same event for one market ticker. Includes research_prompt, a brief for researching the market with web search or a browser tool.
Inputs: ticker (required).
get_market_evidenceGet historical evidence for a marketARead-only
How past events in the same Kalshi series settled: how often markets priced near this market's current price at the same time before close settled YES, how often the favorite won, and the most recent events. Every rate comes with its sample size and a 95% range. Use it for questions like "will this go up?" or "is this likely?", and explain the numbers instead of predicting.
Inputs: ticker (required), events, price_band, hours_before_close.
get_watchlistGet the extension watchlistBRead-only
Current prices for the markets saved in the Chrome extension watchlist.
Inputs: no named parameters.
list_alertsList condition alertsCRead-only
Recent condition matches recorded by the Chrome extension's background monitor.
Inputs: limit, unread_only.
list_saved_screensList saved screensBRead-only
Screens saved through this server plus condition presets and the watchlist shared by the Chrome extension.
Inputs: no named parameters.
run_saved_screenRun a saved screenBRead-only
Run a screen saved through this server or a condition preset shared by the Chrome extension, by name or id.
Inputs: name (required), limit.
save_screenSave a screenAIdempotent
Save search filters under a name so they can be run again with run_saved_screen.
Inputs: name (required), filters (required), description.
search_marketsSearch Kalshi marketsCRead-only
Screen open Kalshi markets by price, time to close, listing age, volume, spread, daily change and event structure.
Inputs: sort, text, limit, scope, +16 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 10 tools
Most tools have clearly distinct purposes: screen CRUD, market search/detail, event detail, evidence, watchlist, and alerts. There is minor overlap because list_saved_screens also mentions presets and the watchlist, while get_watchlist and list_alerts cover related extension data, but descriptions make the boundaries mostly clear.
All tool names use consistent snake_case with a predictable verb_noun pattern such as get_market, search_markets, save_screen, and delete_screen. The verbs are standard and readable throughout.
Ten tools is well-scoped for a market screener and research server. Each tool appears to earn its place by covering a distinct part of the screening, saved-screen, market-detail, or monitoring workflow.
The surface covers the core lifecycle well: save/list/run/delete screens, search markets, inspect markets/events, retrieve evidence, and read watchlist/alerts. Minor gaps exist, such as no update/rename operation for saved screens and no alert management beyond listing recent matches.
Maintenance
Related MCP Connectors
Read-only prediction-market data across Kalshi, Polymarket, Limitless: venues, fees, lookups, stats.
Hosted MCP for Kalshi prediction markets: search, odds, order books, settlement rules, and trading.
Kalshi MCP — US-regulated prediction-market data (no auth on public reads).
Polymarket in your AI: market rules, real fill costs, who is trading, top traders. Read-only.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceProvides access to Kalshi prediction market data with 23 tools for querying events, markets, trades, forecasts, candlesticks, and live data from the CFTC-regulated exchange for trading on real-world events.7-
- AlicenseNot gradedqualityBmaintenanceProvides access to US-regulated prediction-market data from Kalshi, enabling queries on market events without authentication.96 npmMIT
- FlicenseNot gradedqualityCmaintenanceLets AI agents query real Kalshi market data by searching markets, retrieving market details, and reading order books, using only public GET requests with no Kalshi account or API keys.-
- AlicenseAqualityCmaintenanceEnables AI assistants to search Kalshi, Polymarket and Limitless prediction markets by keyword, venue, category or status, inspect series and market details, and review recorded YES bid/ask price history over a market's life. It also reports recording coverage and lists datasets, letting users browse or download daily order book Parquet files with an API key.1040 PyPI1MIT


