Copy-trading on Kalshi lets you mirror the positions of a profitable trader or AI agent without building your own strategy. Instead of spending months developing a trading model, you rent access to someone else’s proven edge and let their trades execute automatically in your account.
This guide walks you through the entire process: finding a reputable copy-trading agent, connecting it to your Kalshi account securely, configuring risk controls, and monitoring performance over time.
What You Will Learn
- How to evaluate and select a copy-trading agent provider for Kalshi
- How to create scoped API keys that limit what the agent can do
- How to configure trade mirroring ratios so the agent sizes positions relative to your bankroll
- How to set daily loss limits and position caps
- How to monitor the agent and know when to disconnect
Prerequisites
Before you start, you need:
- A funded Kalshi account. You must complete KYC verification and deposit at least $100 (recommended minimum $500 for meaningful copy-trading). See the Kalshi agents platform guide for account setup details.
- API access enabled. Kalshi provides API keys through your account settings. You need at least one API key with trade permissions.
- A copy-trading provider. You should have identified 2-3 candidates from the prediction market agent marketplace or the best prediction market bots rankings.
- Basic risk management understanding. Know how much of your total bankroll you are willing to allocate, and what a daily loss limit means for your situation.
Step-by-Step Instructions
Step 1: Research and Shortlist Copy-Trading Agents
Start by identifying providers that support Kalshi specifically. Not all prediction market agents support Kalshi — many are Polymarket-only. Check the agent marketplace for listings that explicitly mention Kalshi support.
For each candidate, evaluate:
- Track record length. Look for at least 3 months of live trading history, not just backtests. Backtested results can be overfit to historical data.
- Strategy transparency. The provider should explain their general approach (event-driven, statistical, news-based) even if they do not share proprietary code.
- Number of active copiers. More copiers suggests more trust, but also means more slippage on entry and exit.
- Fee structure. Subscription vs. profit-share. Profit-share aligns incentives better but can be more expensive if the agent performs well.
Use the bot verification guide to validate any claimed performance metrics.
Step 2: Create a Scoped Kalshi API Key
Never give a third-party agent your main account credentials. Instead, create a dedicated API key with limited permissions.
Log into your Kalshi account, navigate to Settings > API Keys, and create a new key with these permissions:
- Trading: Enabled (the agent needs this to place orders)
- Portfolio read: Enabled (the agent needs to see positions for sizing)
- Withdrawals: Disabled (never grant withdrawal access to a third party)
- Account settings: Disabled
Save the API key ID and secret securely. You will enter these into the copy-trading provider’s dashboard.
Step 3: Verify API Connectivity
Before handing your key to the provider, verify it works. Run this quick check with Python:
import requests
KALSHI_API_BASE = "https://api.elections.kalshi.com/trade-api/v2"
API_KEY_ID = "your-api-key-id"
API_SECRET = "your-api-secret"
# Authenticate and check balance
session = requests.Session()
login_resp = session.post(
f"{KALSHI_API_BASE}/login",
json={"email": API_KEY_ID, "password": API_SECRET}
)
if login_resp.status_code == 200:
token = login_resp.json()["token"]
session.headers.update({"Authorization": f"Bearer {token}"})
balance = session.get(f"{KALSHI_API_BASE}/portfolio/balance")
print(f"Connected. Available balance: ${balance.json()['balance'] / 100:.2f}")
else:
print(f"Auth failed: {login_resp.status_code}")
If this returns your balance, the key is working. If it fails, double-check your key permissions and ensure API access is enabled on your account.
Step 4: Connect the Agent and Configure Mirroring
Log into your copy-trading provider’s dashboard and enter your Kalshi API credentials. Most providers will ask you to configure:
- Capital allocation. The total amount the agent can trade with. Set this to a fixed dollar amount, not “all available balance.” Start with 20-30% of your total Kalshi balance.
- Trade mirroring ratio. If the lead trader risks 5% of their bankroll per trade and you set a 0.5x ratio, the agent will risk 2.5% of your allocated capital per trade.
- Market filters. Some providers let you restrict which Kalshi event categories the agent can trade (politics, economics, weather, etc.). If you are uncomfortable with certain categories, filter them out.
- Max concurrent positions. Limit how many open positions the agent can hold simultaneously. A good starting value is 5-10.
Step 5: Set Risk Controls
This is the most important step. Configure these limits before the agent places its first trade:
- Daily loss limit. If the agent loses more than X% of allocated capital in a single day, it stops trading until the next day. Start with 5-10%.
- Maximum position size. No single trade should exceed a fixed percentage of your allocated capital. Start with 10-15%.
- Drawdown halt. If the agent’s cumulative P&L drops below a threshold (e.g., -20% from peak), pause the agent entirely and review performance.
If your provider does not support these controls natively, you can implement a monitoring script that revokes the API key when thresholds are breached.
Step 6: Paper-Trade for One to Two Weeks
Most reputable copy-trading providers offer a simulation mode or delayed mirroring. Use it. Let the agent generate signals and track what trades it would make in your account for 1-2 weeks before enabling live execution.
During this period, compare:
- The agent’s simulated P&L against its advertised historical performance
- The number and frequency of trades (does it match what was promised?)
- Slippage estimates (the difference between the signal price and the price you would actually fill at)
If the paper results look reasonable, move to live trading. If they diverge significantly from advertised performance, investigate before committing real capital.
Step 7: Go Live and Monitor Weekly
Enable live trading with your configured risk controls in place. For the first month, check the agent’s performance at least twice per week:
- Review total P&L, win rate, and average trade size
- Check that risk limits have not been breached
- Verify the agent is trading the market categories you expected
- Compare actual fills against the lead trader’s fills to measure slippage
Set a calendar reminder to do a full performance review after 30 days. If the agent has not met your minimum return threshold or if drawdowns exceed your comfort level, disconnect the API key and evaluate alternatives.
Common Mistakes and How to Avoid Them
Over-allocating capital. Do not give the agent access to your entire Kalshi balance. Start with 20-30% and increase only after a full month of live performance.
Ignoring slippage. The lead trader gets filled first. By the time the agent mirrors the trade into your account, the price may have moved. This is especially problematic on low-liquidity Kalshi markets. Stick to markets with at least $50,000 in daily volume.
Granting withdrawal permissions. Never. There is no legitimate reason a copy-trading agent needs withdrawal access.
Chasing past performance. An agent that returned 40% last month might have taken outsized risks. Look at risk-adjusted returns (Sharpe ratio, max drawdown) rather than raw percentage gains.
Not setting a drawdown halt. Without a circuit breaker, a malfunctioning or poorly performing agent can drain your allocated capital before you notice. Always set an automatic pause at -20% or less.
Cost Breakdown
| Cost Category | Typical Range | Notes |
|---|---|---|
| Agent rental (subscription) | $75-300/month | Fixed cost regardless of performance |
| Agent rental (profit-share) | 10-25% of profits | Only pay when the agent makes money |
| Kalshi trading fees | $0.01-0.07 per contract | Varies by market and trade size |
| Minimum recommended capital | $500 | Enough for meaningful position sizing |
| Total first-month cost | $100-400 | Including rental, fees, and initial capital |
Profit-share models are more expensive when the agent wins, but they protect you from paying for an underperforming agent. Subscription models are cheaper if the agent performs well but cost you money even during losing periods.
Next Steps and Related Guides
- How to Choose the Right Prediction Market Bot — If you are still deciding between copy-trading and other strategies, use our decision framework.
- How to Set Up a Trading Bot on Kalshi — If you want to build your own bot instead of renting, start here.
- Prediction Market Bot Pricing — Detailed breakdown of pricing models across the ecosystem.
- Prediction Market Bot Verification — How to verify an agent’s claimed performance before committing capital.
- Kalshi Agents Platform Guide — Everything you need to know about running agents on Kalshi.
- Agent Rental Business Guide — Thinking about the rental market from the provider side.