# LLM HyperTracker Skill The **HyperTracker Skill** lets you connect your favourite LLM — ChatGPT, Claude, Cursor, Gemini, Copilot, or similar — directly to HyperTracker’s Hyperliquid intelligence layer. Instead of manually reading API docs, writing endpoint calls, handling parameters, and interpreting raw data, you can simply ask your LLM questions in plain English. For example: > “What are Money Printers doing on ETH right now?”\ > “Find the best-performing wallets trading BTC over the last 30 days.”\ > “Pull this wallet’s closed trades and summarise its strengths and weaknesses.”\ > “Build me a simple dashboard showing smart money positioning by coin.” The skill gives your LLM the context it needs to understand HyperTracker’s API, endpoints, cohorts, data structure, and example workflows. In plain English: **it turns your LLM into a Hyperliquid analyst.** {% embed url="" %} The short video above walks through the full setup in under 60 seconds. It shows how to: 1. Get your free HyperTracker API key 2. Open your preferred LLM or AI coding tool 3. Install the HyperTracker Skill using GitHub or npm 4. Start asking for Hyperliquid insights, dashboards, backtests, wallet analysis, and trading tools Once installed, your LLM can help you pull HyperTracker data, explain what it means, and even generate the code used to fetch it. ### Why this matters HyperTracker gives your LLM access to intelligence that is not available from the raw Hyperliquid API alone. HyperTracker tracks and pre-computes data across: * Every Hyperliquid perp wallet * Open positions * Closed trades * Trader cohorts * Smart money flows * Whale positioning * Liquidation risk * Market bias * Wallet leaderboards * Behavioural analytics That means you can ask higher-level questions instead of manually stitching raw data together yourself. Raw API question: > “Call endpoint X, parse response Y, filter by Z, then calculate the long/short ratio.” HyperTracker Skill question: > “What is smart money doing on ETH right now?” Much better. *** ### Who is this for? You do **not** need to be a developer to start using the skill. It is useful for: * Traders who want faster market intelligence * Builders creating dashboards or bots * Analysts researching wallets and cohorts * Content creators looking for data-backed market insights * Founders testing trading product ideas * Vibe coders who want to build without manually reading API docs You can start by asking simple questions in normal language. Your LLM can then explain the data, write the API calls, generate scripts, build dashboards, or help you turn the output into a trading workflow. *** ### Step 1: Get your free API key Create a free HyperTracker account and generate your API key: [**https://hypertracker.io**](https://hypertracker.io) Free accounts include **100 requests per day**, so you can test the skill, explore the data, and start building before upgrading. *** ### Step 2: Open your LLM or AI coding tool The HyperTracker Skill can be used with tools like: * Claude * ChatGPT * Cursor * Gemini * GitHub Copilot * OpenAI Codex / Agents * Other LLMs that support custom instructions, project files, or agent rules The setup differs slightly depending on the tool, but the goal is always the same: > Give your LLM the HyperTracker Skill file so it understands how to use the HyperTracker API. *** ### Step 3: Install the skill The GitHub repository is here: [**https://github.com/Coin-Market-Man/hypertracker-skills**](https://github.com/Coin-Market-Man/hypertracker-skills?utm_source=chatgpt.com) The package includes ready-made skill files for different AI tools, including Claude Code, Cursor, GitHub Copilot, OpenAI Agents, and manual LLM setups. The repository also includes endpoint references, cohort definitions, authentication details, code patterns, example prompts, rate limit guidance, and troubleshooting notes. #### Claude Code ``` npx skills add Coin-Market-Man/hypertracker-skills ``` #### Cursor ``` npx @coinmarketman/hypertracker-skills --cursor ``` #### GitHub Copilot ``` npx @coinmarketman/hypertracker-skills --copilot ``` #### OpenAI Codex / Agents ``` npx @coinmarketman/hypertracker-skills --agents ``` #### ChatGPT, Gemini, DeepSeek, Qwen, or other LLMs For tools that do not support direct installation, use the manual version. Open the file: ``` hypertracker-skill-generic.md ``` Then paste its contents into your LLM’s custom instructions, project instructions, system prompt, or equivalent setup area. *** ### Step 4: Start asking questions Once the skill is installed, start simple. Try prompts like: ``` Use the HyperTracker API to show me what Money Printers are doing on BTC right now. ``` ``` Find the top-performing ETH traders over the last 30 days and summarise what they have in common. ``` ``` Pull the closed trades for this wallet and explain its trading style, win rate, average hold time, best coins, and biggest weaknesses. ``` ``` Build me a simple market dashboard using HyperTracker data. Include smart money bias, whale positioning, liquidation clusters, and top trader activity. ``` ``` Backtest a simple strategy based on wallets that are profitable on both BTC and ETH. ``` Your LLM should be able to tell you what data it needs, which endpoints to call, how to call them, and how to interpret the results. *** ### What you can build with the HyperTracker Skill The skill is designed to help you move from idea to working prototype quickly. You can use it to build: #### Smart money dashboards Track what profitable wallets are doing across different markets. Example: > “Create a dashboard showing the long/short bias of Money Printers, Smart Money, and Whales across BTC, ETH, SOL, and HYPE.” *** #### Wallet research tools Analyse individual wallets and understand how they trade. Example: > “Analyse this wallet’s closed trades and tell me its best market, worst market, average hold time, win rate, profit factor, and biggest behavioural mistake.” *** #### Copy trading research Find wallets worth watching or copying. Example: > “Find wallets with strong 30-day PnL, high win rate, and consistent ETH performance. Exclude wallets with very low trade counts.” *** #### Market bias tools Understand which side different cohorts are positioned on. Example: > “Show me whether whales, Money Printers, and Giga-Rekt traders are currently long or short on BTC.” *** #### Liquidation and risk dashboards Build tools around liquidation proximity and market stress. Example: > “Create a liquidation risk dashboard for BTC showing where open positions are most vulnerable.” *** #### Backtesting workflows Test simple ideas using closed trade and wallet data. Example: > “Backtest what happens when I follow wallets that are profitable on ETH, trade frequently, and have a win rate above 55%.” *** #### Trading bots and alerts Prototype systems that react to HyperTracker intelligence. Example: > “Write a script that alerts me when Money Printers become more than 70% long on HYPE.” *** ### A simple beginner prompt Use this if you are completely new: ``` I am new to HyperTracker and APIs.Use the HyperTracker Skill to help me explore Hyperliquid market intelligence.First, explain what data is available in simple terms.Then suggest 5 useful things I can ask you to build or analyse.Do not assume I am a developer. ``` *** ### A useful builder prompt Use this if you want your LLM to actually start building: ``` You have access to the HyperTracker Skill.I want to build a simple Hyperliquid intelligence dashboard.Start by asking me what markets, cohorts, and timeframes I care about.Then suggest the best HyperTracker endpoints to use.After that, generate the code step by step and explain what each part does. ``` *** ### A trading research prompt Use this if you want insights rather than code: ``` Use HyperTracker data to help me understand the current market.Focus on BTC, ETH, SOL, and HYPE.Summarise:1. What smart money is doing2. What whales are doing3. Which side looks crowded4. Which markets look most interesting5. What risks I should be aware ofExplain everything in plain English. ``` *** ### The main idea The HyperTracker Skill removes the hardest part of working with trading data. You do not need to memorise endpoints.\ You do not need to understand every response field.\ You do not need to start from a blank code file. You can just ask your LLM what you want to know or build. HyperTracker provides the data.\ The skill teaches your LLM how to use it.\ You get answers, code, dashboards, and workflows in minutes. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.coinmarketman.com/llm-hypertracker-skill.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.