← Back to Skills Marketplace
120
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install kalshi-odds-scanner-pro
Description
Real-time scanner comparing Kalshi odds to 6 sportsbooks, auto-buys 8%+ edge plays with Kelly sizing on NBA, NCAAB, NHL, and MLB markets.
README (SKILL.md)
Kalshi Odds Scanner Pro
Compare Kalshi prediction market prices vs 6 major sportsbooks in real-time. Fires automatically on 8%+ edge. Kelly-sized execution. The exact scanner used to deploy capital daily on Kalshi sports markets.
💰 Used to generate consistent returns on Kalshi sports markets. $79 value.
What It Does
- Fetches live odds from The Odds API (6+ sportsbooks: DraftKings, FanDuel, BetMGM, Caesars, etc.)
- Compares sportsbook-implied probabilities vs Kalshi ask prices
- Fires on 8%+ edge (YES side) or 5%+ edge (NO side heavy favorites)
- Kelly criterion position sizing (25% fractional Kelly, capped at $60)
- NCAAB heavy-favorite NO-side insight: ~74% historical win rate when fav > 80%
- Deduplicates — ONE side per game only
Setup
- Copy
odds_scanner.pyto your polymarket/trading directory - Get a free API key at the-odds-api.com
- Set your Kalshi API credentials:
KALSHI_KEY_ID— your Kalshi API key ID~/.config/kalshi/private_key.pem— your Kalshi private key
Edit constants at the top of the script:
ODDS_API_KEY = "your_key_here"
KALSHI_KEY_ID = "your_kalshi_key_id"
Usage
# Scan YES plays (default NBA)
python3 odds_scanner.py
# Scan NO plays (heavy favorites, 74% win rate)
python3 odds_scanner.py --side no
# Scan both YES and NO
python3 odds_scanner.py --side both
# Scan NCAAB (college basketball)
python3 odds_scanner.py --sport ncaab --side both
# Execute found plays on Kalshi
python3 odds_scanner.py --buy --sport nba --side both
# Set custom edge threshold
python3 odds_scanner.py --min-edge 0.10
Supported Sports
| Key | League |
|---|---|
nba |
NBA Basketball |
ncaab |
NCAA Basketball |
nhl |
NHL Hockey |
mlb |
MLB Baseball |
Edge Logic
YES side: sportsbook_prob - kalshi_yes_ask > 8%
- Example: Sportsbooks say Lakers win 72%, Kalshi YES at 62% → +10% edge → BUY
NO side: (1 - sportsbook_prob) - kalshi_no_ask > 5%
- Example: Sportsbooks say team wins 85%, Kalshi NO at 8% → true NO worth 15% → +7% edge → BUY NO
Kelly Sizing
f = (b*p - q) / b × 0.25 (quarter Kelly)
MIN_BET = $10,MAX_BET = $60RESERVE = $50kept aside always
Integration
Works with ensemble.py and momentum.py in the same directory for multi-model consensus gating.
Requirements
- Python 3.9+
cryptographylibrary:pip install cryptography- The Odds API key (free tier: 500 requests/month)
- Kalshi account with API access
Usage Guidance
This script will read a Kalshi private key from ~/.config/kalshi/private_key.pem and call external APIs, and it can place live trades. Before using: 1) Do not run with --buy until you audit the code and test in dry-run mode; run only scans first. 2) Remove any hardcoded API keys or verify their provenance — the bundled ODDS_API_KEY and KALSHI_KEY_ID in the script are unexpected; treat them as suspicious. 3) Confirm the Kalshi endpoint (api.elections.kalshi.com used in code) is correct for sports trading — if it is wrong or unusual, stop and ask the author. 4) Be cautious integrating ensemble.py/momentum.py from untrusted sources; those local modules can execute arbitrary code. 5) Prefer using a limited/test Kalshi account or sandbox keys if available, and back up your private key before testing. 6) Ask the publisher for provenance (homepage, source repository, author identity, changelog) and for a dry-run / verbose-only mode to validate behavior without transacting. If the author provides a verified repo and explains the hardcoded keys are placeholders, confidence could be raised.
Capability Analysis
Type: OpenClaw Skill
Name: kalshi-odds-scanner-pro
Version: 1.0.0
The skill bundle contains a functional Kalshi trading bot that performs automated financial transactions. It is classified as suspicious due to the presence of hardcoded API credentials (ODDS_API_KEY and KALSHI_KEY_ID) in odds_scanner.py, which is a significant security vulnerability. Additionally, the script accesses sensitive local files (~/.config/kalshi/private_key.pem) to sign requests and executes trades based on external data from The Odds API, posing a high financial risk. While the behavior aligns with the stated purpose, the combination of hardcoded secrets and automated financial execution warrants caution.
Capability Assessment
Purpose & Capability
The name/description match the code: the script fetches sportsbook odds, compares to Kalshi prices, computes Kelly sizes, and can place buys. However the registry metadata declared no required credentials or env vars while both SKILL.md and the code clearly require Kalshi credentials and an Odds API key — this mismatch is unexpected and incoherent.
Instruction Scope
SKILL.md instructs editing constants in the script and placing a private key at ~/.config/kalshi/private_key.pem. The runtime instructions and code will read that private key and sign/submit trade requests to Kalshi. The SKILL.md also suggests integrating optional local modules (ensemble.py, momentum.py) which, if present, will be imported and influence decisions — this expands scope to arbitrary local code. The script contacts two external APIs (The Odds API and a Kalshi endpoint) and can execute real trades (when run with --buy). There are no instructions to run in dry-run mode by default, so a user could accidentally execute live trades.
Install Mechanism
This is an instruction-only skill with a single Python file and no install spec; risk from install mechanism is low (nothing automatically downloaded or written). It does require the 'cryptography' Python package per SKILL.md which is a normal dependency.
Credentials
The registry metadata lists no required env vars, but the SKILL.md and code require (and expect) a Kalshi private key file and an Odds API key. The code includes hardcoded credentials (ODDS_API_KEY and KALSHI_KEY_ID) embedded in odds_scanner.py — hardcoded keys are suspicious and inconsistent with instructions that say 'get your own API key' and set constants. Requesting access to a private key file in the user's home directory is proportionate to signing trade requests, but it should have been declared explicitly in metadata; the undeclared/embedded credentials and mismatch are red flags.
Persistence & Privilege
always is false and there's no install step that forces persistence or modifies other skill configurations. Autonomous execution (model invocation) is allowed by default but not exceptional here; combined with credential access and trading capability this increases potential impact if misused, so the user should be cautious about granting agent autonomy.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install kalshi-odds-scanner-pro - After installation, invoke the skill by name or use
/kalshi-odds-scanner-pro - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Frequently Asked Questions
What is Kalshi Odds Scanner Pro?
Real-time scanner comparing Kalshi odds to 6 sportsbooks, auto-buys 8%+ edge plays with Kelly sizing on NBA, NCAAB, NHL, and MLB markets. It is an AI Agent Skill for Claude Code / OpenClaw, with 120 downloads so far.
How do I install Kalshi Odds Scanner Pro?
Run "/install kalshi-odds-scanner-pro" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Kalshi Odds Scanner Pro free?
Yes, Kalshi Odds Scanner Pro is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Kalshi Odds Scanner Pro support?
Kalshi Odds Scanner Pro is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Kalshi Odds Scanner Pro?
It is built and maintained by Mike (@themsquared); the current version is v1.0.0.
More Skills