KellyIQ

com.kellyiqPublished by kellyiq.comOfficial publisherv1.0.0Updated Oct 30, 2026

Size an entire sports betting slate as one portfolio, using fractional Kelly and exposure limits.

VerifiedStreamable HTTPWeb executableFinanceData & Analytics

Overview

AI-generated overview

Hosted KellyIQ MCP server that lets an assistant size a whole sports betting slate as one portfolio using fractional Kelly and exposure limits.

What you can do
  • Size a slate: send a bankroll and the bets you have chosen with your own probability view, and get a dollar stake for every bet.
  • Apply margin removal and limits: prices are de-vigged, your view is applied, and stakes are sized together under per-bet and total-exposure limits.
  • Repeat results: the response includes a seed so the same answer can be reproduced.
  • Explain the outcome: a bundled skill guides the assistant to collect the slate, confirm before optimizing, and explain stakes and risk.
When to use it
Useful when you already have your own picks and probabilities and want an assistant to turn them into consistent, portfolio-level stake sizes instead of sizing each bet in isolation. It does not choose bets or predict outcomes.
Requirements
Remote hosted endpoint at api.kellyiq.com over streamable HTTP; no local package needed. Authentication is required: OAuth sign-in for OAuth-capable clients such as Claude, or a KellyIQ API key sent in the Authorization header as a Bearer token. Keys are created in the KellyIQ app under Settings, and every plan including Free has a daily allowance.
Before you install
The Authorization header carries a secret API key; keep it out of committed or shared code and prefer OAuth sign-in where available. Requests send your bankroll and bet details to a third-party hosted service. Results are calculations, not betting advice, and the service sizes bets you choose rather than picking them; usage is subject to plan allowances and rate limits, and it is intended for adults where legal.

Installation

In SourceWeft

  1. Open KellyIQ in the dashboard and add it to a workspace.
  2. Connect the required credentials. They are stored privately and never shown on public pages.
  3. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/com.kellyiq%2Fkellyiq/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.

{
  "mcpServers": {
    "kellyiq": {
      "type": "http",
      "url": "https://api.kellyiq.com/mcp"
    }
  }
}

README

KellyIQ examples

Working examples for the KellyIQ API: size a whole sports betting slate as one portfolio from your own scripts, models, spreadsheets and AI assistants.

You send your bankroll and the bets you have chosen, with your view of each. KellyIQ removes the bookmaker's margin from each price, applies your view, and returns a dollar stake for every bet, sized across the slate together under per-bet and total-exposure limits.

Get a key

Create an API key in the KellyIQ app under Settings → API. Every plan includes API access, Free included, each with its own daily allowance. Then:

bash
export KELLYIQ_API_KEY="kiqa_..."

Recipes

Each recipe is a complete script in Python (requests) and JavaScript (Node 18+, no packages).

RecipeWhat it shows
first-optimizeSize a three-bet slate and read every part of the result
slate-from-csvKeep your slate in a spreadsheet: CSV in, CSV of stakes out (slate.csv)
reproducible-resultsGet exactly the same answer again with the returned seed
handling-limitsBack off and retry on rate limits and a busy server
bash
cd recipespython first-optimize.py        # or: node first-optimize.mjs

Each recipe has a walkthrough in the docs: Recipes.

Python package

pip install kellyiq (kellyiq-python): the same API with specific exceptions for each error and automatic retries when the server is busy.

ExampleWhat it shows
python/size-a-slate.ipynbA notebook: stakes, the probabilities behind them, and the risk at each Kelly fraction, as tables
python/slate_from_csv.pyCSV in, CSV of stakes out, with each error handled (slate.csv)
python/size_from_model.pyYour model's probabilities in, stakes merged back into the DataFrame (model_output.csv); needs pandas
python/nba_home_teams.pyA real NBA slate (DraftKings prices, Oct 30, 2026) sized with every home team at the same assumed Moderate Edge (nba_2026-10-30.csv); needs pandas

Guide: Python SDK.

Excel

spreadsheets/kellyiq-slate.xlsx: enter your bankroll and your bets, press Ctrl+Alt+F5, and get a stake for every bet with a summary of the slate. No macros, and your API key stays in a file on your computer, never in the workbook. Excel for Windows. Setup: spreadsheets/.

Benchmark

benchmarks/slate-sizing: three NBA slates sized by cap-then-scale, by the exact constrained optimum and by KellyIQ, scored by exact expected log growth. The slates, the scoring script and every stake are included so you can rerun it.

AI assistants (MCP)

KellyIQ runs a hosted MCP server at https://api.kellyiq.com/mcp, so your assistant can size a slate as one of its tools. Claude connects by signing in; coding tools use an API key, with ready-to-use configs in mcp/:

AssistantSetup
Claude (web, desktop, mobile)Settings, Connectors, Add custom connector, paste the URL, then Connect and sign in
Claude Codemcp/claude-code.md
Cursormcp/cursor.json
VS Code (GitHub Copilot)mcp/vscode.json
Gemini CLImcp/gemini-cli.md

Full guide: AI assistants (MCP).

Skills

skills/analyze-my-slate teaches the assistant to collect your slate in one message, confirm before using an optimize, and explain the stakes and the risk. Install steps for Claude and Claude Code: skills/.

Docs and support

Keep your key out of code you commit or share; every example here reads it from the environment.

KellyIQ sizes the bets you choose; it does not choose bets or predict outcomes. Results are calculations, not betting advice. 21+, where legal.

The examples are MIT licensed. The KellyIQ optimizer itself runs as a hosted service.

Source: README.md at commit 27fb178

Tools

0
This server asks you to sign in, so SourceWeft could not read its tools without an account and has not indexed them. What a connected account sees can depend on that account and its permissions.

Version history

1
  1. v1.0.0LatestOct 30, 2026

Score

85% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts