Why it matters
This project offers AI builders a framework to develop and deploy sophisticated trading strategies directly from AI models. It enables automated execution of trades with precise risk management, bridging the gap between AI decision-making and real-world financial markets.

What changed

The Pathiel project introduces an AI trading agent that connects to the Hyperliquid perpetual futures exchange. This agent functions as an MCP (Multi-Client Protocol) server and a Claude Code plugin, enabling AI models like Claude to interact with market data and execute trades. It supports both research-only modes, which only read market data and account information without placing orders, and full trading modes. For trading, Pathiel requires a Hyperliquid API wallet and automatically sizes trades based on a user-defined risk profile, defaulting to 3x leverage and 1% of equity per trade. Each order includes an exact stop-loss and take-profit, forming a bracket order. The agent provides desk briefs for various markets, including crude oil, metals, and the S&P 500, offering AI models written calls to inform their trading decisions. Installation is streamlined, with options for direct integration into Claude Code via plugin marketplace commands or through shell commands and Docker containers for broader MCP client compatibility. The project is released under version v0.9.0 and is written in Python.

Why it matters for builders

Pathiel provides AI developers with a concrete example of how to integrate AI agents into high-frequency trading environments. It demonstrates a practical application of AI for financial markets, showcasing how complex decision-making and order execution can be automated. The project's open-source nature and clear integration paths via MCP and SDKs allow builders to experiment with, adapt, and extend its capabilities for their own trading bots or AI-driven financial tools.

Practical impact

Builders can leverage Pathiel to develop AI trading strategies that execute trades with precise risk controls directly from their AI models. The project's integration with Claude Code and the Claude Agent SDK means that developers can start experimenting with trading prompts and see how the agent responds. For those with a Hyperliquid account, the setup process is designed to be quick, allowing for the testing of real-money trades with a default risk configuration. The availability of desk briefs also offers a structured way to feed market analysis into AI decision-making processes. Developers can explore the provided prompts to understand the agent's capabilities, from retrieving market data and account information to sizing and placing bracket orders.

Caveats and source limits

This analysis is based on the provided GitHub repository information and excerpt. Specific details regarding independent performance benchmarks, real-world trading results, or advanced customization options beyond the described setup are not detailed. The excerpt mentions a "fresh release" and version v0.9.0, indicating that the project may still be under active development and subject to changes. While it supports trading on Hyperliquid mainnet, users are advised to practice on testnet first. The exact parameters for the "3x / 1% risk by default" can be configured, but specific default values beyond this description are not enumerated. The excerpt does not provide pricing information, as the project is open-source and free.

Sources

Written with AI assistance from the linked sources; every claim below was checked against them automatically. How we produce articles.

Claim check: 7/7 supported claims - 7 evidence links - 100% avg confidence
  • Pathiel is an AI trading agent for Hyperliquid that trades perpetual futures.supported - github.com
  • The agent integrates with Claude Code and other MCP clients.supported - github.com
  • Pathiel places orders with an exact stop and target for every trade.supported - github.com
  • The default risk profile is 3x leverage and 1% of equity per trade.supported - github.com
  • Pathiel provides desk briefs for oil, metals, and S&P 500.supported - github.com
  • The project is free and open-source.supported - github.com
  • The latest release is v0.9.0.supported - github.com

Caveats

  • Single-source caution: verify critical details at the linked source.
Radar score 87/100 - how it was calculated
Reliability82
Freshness100
Novelty77
Technical85
Developer96
Ecosystem72
Confidence96
  • Reliability 82: GitHub metadata supports source trust
  • Freshness 100: Fresh GitHub release date
  • Novelty 77: Fresh GitHub release
  • Technical 85: Repository technical metadata
  • Developer 96: Developer tooling signals
  • Ecosystem 72: Fresh GitHub release
  • Confidence 96: Claims have reliable evidence
Share
XLinkedInHacker News

Discussion

Loading comments...

Related articles

Agents - Nov 22, 2026Waku Agent: Open-Source AI Agent Harness with Local-First ArchitectureWaku Agent is an open-source AI agent harness designed for local-first operation, featuring a readable codebase that emphasizes core agent pillars: harness, loop, memory, eval/LLM-Ops, and an always-on layer. It offers local data storage, a straightforward Python loop, and integrated evaluation tools.Agents - Nov 22, 2026matthewdonsemail-lab/log: AI Agent Social Media Automation CLIThe matthewdonsemail-lab/log project offers an open-source CLI and MCP server designed for AI agents like Claude Code and Codex. It enables automated management of social media interactions across platforms such as X, Discord, Slack, Instagram, Reddit, and LinkedIn, operating in the user's distinct voice.Agents - Nov 22, 2026AgentForge: Open-Source AI Coding Agent HarnessAgentForge is an open-source terminal-based harness for AI coding agents, designed for studying agent loops, tools, and persistence. It supports multiple AI models and provides a framework for developing and testing agent behaviors.Agents - Nov 21, 2026Siyuan v3.8.11-rc.3: AI Agent Knowledge WorkspaceSiyuan has released version v3.8.11-rc.3 of its open-source, privacy-first, self-hosted knowledge workspace. This update focuses on enabling humans and AI agents to collaborate within the platform.Agents - Nov 21, 2026Fetch.ai Releases 80+ AI Agent ExamplesFetch.ai has released a repository containing over 80 production-ready AI agent examples in Python. These examples cover autonomous agents, multi-agent systems, and agentic AI workflows, utilizing frameworks like uAgents, ASI:One, MCP, LangChain, and CrewAI.Agents - Nov 22, 2026WeChat DevTools MCP Server and Skill ReleasedA new version of the MCP server and skill for WeChat DevTools has been released, enabling AI-powered compilation, preview, debugging, and auto-testing of Mini Programs. This tool supports both Windows and macOS environments.