$SERV
StableSnapshot Window: 2026-07-09 14:30 UTC · ← Back to Crypto Overview
Social Momentum Summary
Total Engagement - Comments: 4, Retweets: 8, Likes: 37, Impressions: 1571
Verbatim Community Citations & Social Evidence 1 source posts analyzed
New on the Limitless Academy: a quickstart for running a market-making bot with an AI agent. It quotes both sides of a market and hedges every fill on a second venue to stay delta-neutral. The agent supervises, you keep custody. Point your agent at the repo, set ~5 vars, go [Spoken audio]: It is 2026. If your job title isn't prediction market market maker using AI agents, what even is your occupation? Permanent underclass behavior. Okay, memes aside, using agents as market makers is an emerging research field. And it's something that we're going to learn how to do today on the Limit List Academy. Whether you just want to improve your prediction market knowledge or upgrade your trading from manual to using an API, academy.limitlist.exchange is the place to do it and today we're gonna head into the Agents Academy where we'll learn how to wire an LLM into using the limitless prediction market and we're gonna use a market making bot quick start to run a strategy that quotes both sides of a market on one venue and hedges every fill on a second venue to stay Delta neutral. We're gonna add an AI agent layer on top to supervise it and we're gonna use poly market and limitless today to get this Delta neutrality and what we'll end up with is a dashboard that looks like this or even if you want to run this in Hermes agent, a harness of your choice, you can. And this is where I have my, where I'm a whole eight cents in profit right now. Pretty serious numbers. And look, if you're comfortable with agentic engineering and vibe coding, you could probably just copy this page as markdown, feed it to your AI agent and get started very easily. The main repository we're going to be working with is the agent starter repo, which I've put together to help people use their AI agents to get started with trading programmatically on limit list very quickly and easily. And my preferred stack still is using cursor with Claude code inside of it. Probably what I recommend to you guys. Now that we're in cursor, it's really as easy as just set me up for market making. And this whole repository is optimized for the agent experience. So really it's just gonna walk you through the basic environment variables that we have to set up and we'll be moving. Essentially all we'll need is the private key which you can get from your wallet extension, usually by going to accounts and export private key. The next thing we'll need is limitless token ID and token secret, which you can generate via limitless.exchange. You can do that in the upper right, click to your profile, go to API key, derive new token, and that'll give you your limitless token ID and token secret. Now this is actually all we'll need for a dry run, but to go to production, if you wanna do this cross-venue poly market limitless thing. The easiest way to get everything funded and going is to get your relayer API key and API key address, which will generate the depositor wallet on poly market for you. Just to show you what that looks like. We can go to poly market and I'll show you the flow. Basically all we've got to do is we go to your profile, client-dazzle, settings, relayer API keys and you can create a key here and essentially it'll help you set up all the necessary wallets and it'll give you the exact wallet to fund for your poly market side of things and you'll be ready for production. Now that the environment is set, we can go ahead and tell it M is set. Let's get this show on the road and your agent harness is going to cook everything else for you. And the reason why you'd use an AI agent here is not only to smooth and out the actual deployment of this spot and the running of it. But if we look at Hermes, for example, what I think is quite interesting is that you're able to generate learnings per iteration of your AIH market maker where it'll come up with some improvements where maybe, hey, you should run this controlled fill rate test next so that we can start exploring different candidate pairs and hedge qualities. So I think it really serves well as a learning and thinking layer on top of your market making. It also just helps people get started. And like I said, the other good thing about this repository is that it's going to use the limitless official SDKs very effectively so have a very smooth experience in running your bot. So we'll let this cook for a second. While we let Fable cook and finish up the deployment of our market-making bot, I also want to highlight a few other resources we've given here. If you want to have some more persistent infrastructure for your trading bot, I can recommend Railway. They've done a lot here to actually optimize for the agent experience. This is a very good place where you can deploy your trading bot, keep it persistent, have it on managed infrastructure. And if you want further guidance on how to actually navigate Limitless, API reference, SDK reference, definitely check out docs.limitlessexchange for all of that. Our fable instance has finished cooking. We've got a nice little dashboard here. This is in fact a dry run, but it's a real hedge that can be filled. Again, to go to production, just fill out those five variables and you're done. Please feel free to ping with any questions. And if you wanna join the discussion, hop into the Limitless Builders Chat on Telegram. That's at Limitless Builders Chat. Definitely come by. Thanks guys for watching.
