Why it matters
This project provides builders with an AI-driven approach to automated trading on Hyperliquid, integrating advanced risk management with AI capabilities. The inclusion of a free Claude Code plugin offers developers a direct way to leverage AI for trading strategy development and execution.

What changed

The Pathiel repository, developed in Python, has recently seen its latest release, version v0.4.1, on November 13, 2026. This release marks a fresh update for the AI trading agent, which is built to operate on the Hyperliquid platform. The agent leverages Anthropic's Claude model to execute trades for perpetual futures. A key feature is its adherence to the MCP (Multi-Collateralized Positions) framework, ensuring that every order is placed with an exact stop-loss and take-profit target. By default, the agent employs a risk management strategy of 3x leverage or a 1% risk per trade, reflecting a structured approach to managing capital. The project also incorporates "oil and metals desk briefs," suggesting a focus on commodity markets. Additionally, a free Claude Code plugin is provided, enhancing its utility for developers.

Why it matters for builders

For developers and traders interested in algorithmic trading, Pathiel offers a concrete example of an AI agent integrated with a specific trading platform (Hyperliquid) and a sophisticated AI model (Claude). The emphasis on precise risk parameters and the inclusion of a dedicated plugin for AI code generation lowers the barrier to entry for those looking to experiment with or build upon AI-driven trading systems. The project's structure, including its Python base and package management files like pyproject.toml and requirements.txt, provides a clear blueprint for similar agent development.

Practical impact

Builders can explore the Pathiel repository to understand the implementation of an AI trading agent that combines LLM capabilities with structured trading protocols. The availability of the Claude Code plugin suggests potential for rapid prototyping and iteration of trading strategies. The project's recent release indicates ongoing development and support. Interested parties can examine the code, particularly the MCP integration and risk management logic, to adapt it for their own trading environments or to contribute to the project. The repository has 36 stars and 10 forks, indicating a level of community interest.

Caveats and source limits

The provided metadata indicates a recent release and a focus on specific trading functionalities. However, details regarding performance benchmarks, specific backtesting results, or comprehensive usage guides are not present in the repository summary. The exact nature and scope of the "oil and metals desk briefs" are not elaborated upon. Furthermore, while the project is open-source under the MIT license, information on community support channels beyond GitHub issues, or detailed documentation beyond the README, is limited. The metadata does not specify the exact version of Claude or the Claude Code plugin used, nor does it detail the specific requirements for running the agent beyond the presence of a Dockerfile and requirements.txt.

Sources

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

Claim check: 10/10 supported claims - 10 evidence links - 100% avg confidence
  • Pathiel is an AI trading agent for Hyperliquid that uses Claude to trade perpetuals via MCP.supported - github.com
  • The agent places orders with an exact stop and target.supported - github.com
  • The default risk setting is 3x leverage / 1% risk per trade.supported - github.com
  • The project includes oil and metals desk briefs.supported - github.com
  • A free Claude Code plugin is available.supported - github.com
  • The project is written in Python.supported - github.com
  • The latest release was on November 13, 2026.supported - github.com
  • The repository has 36 stars.supported - github.com
  • The repository has 10 forks.supported - github.com
  • The project is licensed under MIT.supported - github.com

Caveats

  • Single-source caution: verify critical details at the linked source.
Radar score 86/100 - how it was calculated
Reliability82
Freshness100
Novelty73
Technical77
Developer96
Ecosystem72
Confidence100
  • Reliability 82: GitHub metadata supports source trust
  • Freshness 100: Fresh GitHub release date
  • Novelty 73: Fresh GitHub release
  • Technical 77: Repository technical metadata
  • Developer 96: Developer tooling signals
  • Ecosystem 72: Fresh GitHub release
  • Confidence 100: Claims have reliable evidence
Share
XLinkedInHacker News

Discussion

Loading comments...

Related articles

Agents - Nov 19, 2026Pi AI Agent Runtime Releases Pluggable Extension PackagesThe Pi AI agent runtime has released a suite of pluggable extension packages, offering LLM-callable tools for various functionalities. These packages cover integrations with IM workspaces like Lark, DingTalk, and WeCom, alongside browser and computer automation, persistent memory, telemetry, task scheduling, image generation, web access, and security policies.Agents - Nov 19, 2026jnMetaCode/agency-agents-zh: 277 AI Experts & OrchestratorThe jnMetaCode/agency-agents-zh repository offers 277 plug-and-play AI expert roles, supporting over 20 tools like Claude Code, Cursor, and Copilot. It includes 64 original agents tailored for the Chinese market across various sectors, and integrates with the agency-orchestrator for DAG-based multi-agent collaboration.Agents - Nov 17, 2026BuilderIO Agent Native: Framework for Agentic AppsBuilderIO Agent Native is an open-source TypeScript framework designed for building agentic applications. It recently released version v0.1.298, indicating active development and a focus on providing tools for AI agents.Agents - Nov 20, 2026Grajo: Self-Hosted Agentic Framework and MCP PlatformGrajo is an open-source, self-hostable agentic framework and Model Context Protocol (MCP) platform. It allows users to build and run AI agents with persistent memory, research capabilities, and P2P sharing, while also serving as an MCP platform to connect existing AI clients like Claude Code and Cursor.Agents - Nov 19, 2026OpenCrawl Core: Multiplayer Context Layer for Coding AgentsOpenCrawl Core introduces a multiplayer context layer designed to enhance collaboration among coding agents. It enables agents to share search results, memory files, and coordinate actions, working with models like Claude Code and Codex.Agents - Nov 18, 2026llmscribe v1.4.1: AI Agent Filesystem Interfacellmscribe has released version 1.4.1, offering a structured interface for AI agents to interact with local project files across Linux, macOS, and Windows. This Python-based tool aims to provide a consistent filesystem layer for AI agents.