Author: Changan | Biteye Content Team
What time did you last stop watching the market?
The longest I've ever watched the market was from the opening of the US stock market until the closing of the Asian market. I slept for two hours in between, and when I woke up, I realized I had missed a perfect entry point.
The editor started to wonder: Why does this task absolutely need to be done by someone?
After integrating OpenClaw with Hyperliquid and Polymarket, I tested the answer to this question. Some things went more smoothly than expected, while some pitfalls were more complex than anticipated. This post covers it all.
Because you got caught in the rain, Biteye wants to hold up an umbrella for you.
I. Installation Steps
If you want AI to trade for you, simply installing OpenClaw isn't enough. The AI itself is only responsible for scheduling; the actual trading is done by the skill. And for the skill to run, you need to feed it the corresponding trading permissions.
This chapter explains the three things—installation, configuration, and push notifications—all at once. After configuring them, you can directly move on to the practical part of Chapter Two.
1.1 Install Skills: Find the capabilities you need
For any task you want to complete, you can search for and install the necessary skill keywords on ClawHub, for example:
hyperliquid skill
polymarket skill
After installation, confirm that skill is loaded and you can access the local console page: Left-hand agent panel → Click skill
If you can see the newly installed skill, the installation was successful.
At this point, some skills will display "Save key," indicating that you need to continue configuring the API or private key. This is normal. The next step is to fill in the corresponding permission information.
1.2 Configure the trading API: Grant the AI order placement permission
To enable AI to place orders, it must be granted trading permissions.
Do not use the main account directly. It is recommended to open a new account for testing and put only a small amount of funds in it.
Different exchanges have different acquisition methods; here are two examples.
1️⃣Hyperliquid
Open Hyperliquid → More → API
This allows you to create an API wallet with transaction permissions but no withdrawal permissions.
Enter your wallet name → Click Generate to obtain the API key, and you can customize the authorization period.
After generating your API wallet, based on my own hard-earned experience, I want to assure you: do not transfer any funds to your API wallet address. Keep your funds in your main account. (Funds transferred to your API wallet will be lost.)
The configuration requires two pieces of information: the main account's wallet address (API Key) and the private key displayed when the API wallet is generated (API Secret). AI operates on the funds in the main account through API wallet signing.
The testing method involves having OpenClaw query your account balance by entering the following in the dialog box: "Query my current balance in Hyperliquid".
If a specific number is returned, it means the API configuration is normal.
2️⃣Polymarket
Open the Polymarket settings page:
Settings → Relayer API Keys → API key →Click Create New
After creation, you will get:
RELAYER_API_KEY
RELAYER_API_KEY_ADDRESS
Signer Address
If you need to conduct a transaction, you also need to export the wallet private key corresponding to the signer and fill it into the configuration.
Here's a demonstration using OKX Wallet: Click your profile picture in the top left corner of your wallet → Manage Wallet → View Private Key
Next, enter the private key and API key into the openclaw.json file of OpenClaw. Then open openclaw.json (the path is usually C:\Users\your username\.openclaw\) and enter the three items of data in the env field.
⚠️ The private key is the highest level of authorization. Do not take a screenshot, do not send it to anyone, and do not save it in your chat history.
After configuring the API, you will need to complete an on-chain authorization. The simplest way is to manually execute a transaction on Polymarket, for example, to buy a position of 1 USDC.
1.3 Connect to Telegram: Use it even when you're out and about.
After configuring the API, crayfish can be traded, but you can't keep staring at the computer screen, right?
After integrating with Telegram, you can handle everything from transaction notifications and error reports to placing an order on a whim with just your phone. Whether you're out for a meal or lying in bed, your crayfish will do whatever it wants, and you'll always know what it's doing.
Create a Telegram bot:
Open Telegram
Search BotFather
To create a Telegram bot: Send /start → /newbot → follow the prompts to create the bot → you will receive a bot token after creation.
To obtain the Chat ID: Send a message to the bot, then open: https://api.telegram.org/bot (replace this part with your token)/getUpdates
In the returned JSON, look for chat → id; that string of numbers is your Chat ID.
Then fill in the two parameters in the configuration:
TELEGRAM_BOT_TOKEN
TELEGRAM_CHAT_ID
Then test: Send me a test message
If you receive the message on Telegram, the push was successful.
If you cannot receive the message, the common cause is a proxy or network problem. You need to fill in the HTTPS proxy port in the configuration.
Open the openclaw.json file and add:
{
"HTTPS_PROXY": "http://127.0.0.1:yourport",
"HTTP_PROXY": "http://127.0.0.1:your port"
}
Enter the port number according to your own proxy software.
II. Order Placement: Turning Judgments into Actions (More Details)
Ever since AlphaGo's match against South Korean Go player Lee Sedol, humans have been particularly enthusiastic about competing against AI.
Last October, an AI research institution called Nof1 held a competition called Alpha Arena. The rules were simple: give each model on the market $10,000 to trade perpetual contracts autonomously on Hyperliquid, and whoever had the highest account balance at the end won.
A while ago, Aster also held two "human vs. AI" live trading competitions, and unexpectedly, AI won both of them.
After reading this, I started to wonder: Could I build a similar pipeline myself and run it on Hyperliquid using OpenClaw?
Thus, the following two skills were created.
2.1 Hyperliquid: Enabling AI to Create Contracts
I searched on ClawHub and found the corresponding skill. For installation instructions, please refer to the deployment tutorial in the previous article.
After configuration, you can place orders directly using natural language: "Go long ETH on Hyperliquid, 10x leverage, 30% position."
But in reality, it wasn't that smooth.
The first problem was leverage. When asked to set a leverage of 10x, it set it to 20x, and even after several attempts to correct it, it couldn't get it right. The expression "10x leverage" was ambiguous for the AI; it would adjust according to its own judgment of "reasonable leverage." Later, a more explicit instruction was given, requiring it to display the specific parameters before placing the order and confirm before execution, which finally stabilized the process.
The second issue was position sizing: when I told Xiaolongxia to "use 30% of the balance," the result was either a full position or an absurdly small position. The problem was only solved when I directly told her the specific amount, such as "open a position of 200 USDC."
My initial goal was to replicate the Alpha Arena model: to make Openclaw run fully automatically, 24/7, without needing to be monitored, just like those AI models.
This approach is theoretically feasible, but you need to first write the strategy in natural language and tell OpenClaw: which instrument, under what conditions to enter, position size, stop-loss order, and closing conditions. Once it understands this, it will continuously monitor and execute according to this logic, without requiring you to manually trigger it each time.
However, the immaturity of the skill ecosystem at this stage is a real hurdle. No matter how good the strategy you feed in is, if there are problems at the execution layer, the results will go astray. Just like the AI model in Arena running on the native Hyperliquid environment, the stability is on a completely different level compared to indirectly calling it through a third-party skill.
This part will be run again once the skill ecosystem matures.
2.2 Polymarket: Applying the same logic to prediction markets
After spending some time on Hyperliquid, I started to wonder: how will OpenClaw perform in the prediction market?
The underlying logic of the two markets is actually quite similar; both are about pricing the "probability of something happening." The information sources are the same, and the judgment framework is similar; only the execution points differ.
In the previous article, I only used PolyClaw to analyze the market without actually placing any orders. There's a psychological barrier between analysis and placing orders; if the analysis is wrong, there's no loss, but if the order is wrong, it's real money. But this time, I want to take it a step further.
So, after filling in the three keys and completing the on-chain authorization according to the configuration process in Chapter 1, I asked OpenClaw to help me find a worthwhile opportunity to enter the market.
It pushed a market indicator: "Will interest rates remain unchanged after the Fed's March meeting?" Currently, YES is quoted at $0.99, with a 24-hour trading volume of $2.4M, and the market consensus is 99% that rates will remain unchanged. Fundamentals also support this judgment: inflation data is stable, economic growth is moderate, and the Fed's recent stance is dovish but not in a hurry to cut rates.
Before placing an order, I usually ask: "What is the current order book depth in this market? How much spread is there? How much should I buy?"
OpenClaw analyzes the current bid-ask spread, and placing a limit order when the spread is too large can save a lot of unnecessary losses.
After confirming that the depth is acceptable, tell Openclaw whether to buy and how much to buy to place the order: "Buy 5 USDC YES position."
Checking your holdings after a transaction is straightforward: "OpenClaw, what positions do I currently hold?"
From discovering an opportunity to placing an order, the entire process is completed within OpenClaw, eliminating the need to switch pages or endure the often laggy Polymarket frontend.
III. Conclusion
The AI models in Alpha Arena run on Hyperliquid with $10,000 on their own, without sleeping, without getting emotional, and without losing their rhythm because of a losing trade.
Previously, this infrastructure was only accessible to professionals with technical backgrounds. Now, OpenClaw has lowered the barrier to entry to the average person. You don't need to know how to code or build your own system; a simple natural language command can have AI execute it for you.
The tools are already here.
There's only one question left: when do you plan to start?
