> For the complete documentation index, see [llms.txt](https://olaxbt-docs.gitbook.io/olaxbt-doc/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://olaxbt-docs.gitbook.io/olaxbt-doc/technical-architecture/ai-market-maker.md).

# AI Market Maker

We’re sharing our [AI Market Maker GitHub Repository](https://github.com/olaxbt/ai-market-maker), providing transparency into the technology powering these capabilities and fostering a collaborative community of innovators. The repository includes key components such as exchange data fetching, price pattern analysis, sentiment tools, technical indicators, and arbitrage strategies, offering a robust foundation for trading innovation.

<figure><img src="/files/rN1evzUktsbenFN6DoZ2" alt=""><figcaption></figcaption></figure>

The customisable trading tools showcase the power of our Model Context Protocol (MCP) framework in empowering AI agents.

**AI-Market-Maker** is an open-source project to simulate a cryptocurrency market-making system using AI agents. It aims to provide liquidity and optimise bid-ask spreads for crypto assets like Bitcoin (BTC) and Ethereum (ETH) on exchanges such as Binance and Coinbase. This is an early-stage project for educational purposes, built with a modular agent-based architecture.

## Goals

* Simulate crypto market making with AI agents.
* Fetch real-time crypto market data and place simulated orders.
* Support future features like arbitrage, sentiment analysis, and risk management. Note: This project is an experimental simulation currently in development.

\
Agents
------

* **Market Scanner Agent**: Scans for newly listed, to-be-delisted, and high-potential tokens with high-momentum price shifts.
* **Technical Analysis Agent**: Analyses price patterns and candlestick data for actionable trading insights.
* **Sentiment Analysis Agent**: Processes KOL tweets and news to gauge market sentiment trends.
* **Statistical Arbitrage Agent**: Calculates correlations and hedge ratios to identify token pairs for arbitrage and hedging.
* **Quantitative Analysis Agent**: Generates trading signals by evaluating momentum and market trends.
* **Valuation Agent**: Assesses token value by benchmarking against similar category assets.
* **Liquidity Agent**: Evaluates circulating supply and on-chain liquidity for efficient trades.
* **Risk Management Agent**: Optimises TP/SL and position limits using reinforcement learning on volatility.
* **Portfolio Management Agent**: Manages portfolio and balance for optimised order placement across DEX and CEX venues.

<figure><img src="/files/Yg4Tu3XU3h3LAvRzeBrN" alt=""><figcaption></figcaption></figure>
