This project provides a traceable "Agent Skill" for analyzing the investment views and methodologies of fund manager Zheng Xi (E Fund). It leverages his complete public statements, a distilled investment framework with verbatim evidence, and comprehensive market fund data. Designed for AI agents and technical users, it enables verifiable Q&A, fund scoring, and comparative analysis, strictly avoiding fabricated information and serving as a research aid.

How It Works

The skill is built on three core, verifiable pillars derived from public disclosures. First, an extensive corpus of Zheng Xi's public statements from 2012-2026, including quarterly reports, manager notes, and interviews. Second, a distilled investment methodology, where each principle is directly supported by his original quotes, allowing for reasoned inference on topics not explicitly covered. Third, real-time data for Zheng Xi's managed funds and a database of approximately 27,000 market-wide funds for comparison and scoring. This multi-source approach ensures all outputs are traceable to original content or explicitly noted as inferences based on his established methods.

Quick Start & Requirements

• Primary Install: For Claude Code, copy the repository contents to ~/.claude/skills/zhengxi-views/. For Tencent WorkBuddy, import the provided skill.yml. Other AI platforms require custom instructions and uploading knowledge files.
• Prerequisites: Python 3.x is required. The requests, beautifulsoup4, and lxml libraries are necessary for data fetching and fund scoring scripts (pip install -r requirements.txt). Core corpus retrieval functions do not require external libraries.
• Resource Footprint: Setup is minimal for core features; data fetching and scoring scripts may require network access and moderate processing.

Highlighted Details

• Traceability: All conclusions are directly linked to original source text (year, publication, quote).
• Comprehensive Data: Covers Zheng Xi's historical and current funds, plus ~27,000 other market funds for detailed comparison and scoring against his framework.
• Cross-Platform Compatibility: Designed to function within various AI environments including Claude Code, WorkBuddy, ChatGPT, Gemini, and Cursor.
• Methodological Inference: Extends analysis beyond direct statements by applying Zheng Xi's distilled investment framework to new topics, clearly demarcating inferred content.

Maintenance & Community

The README does not detail specific contributors, ongoing maintenance plans, or community channels like Discord or Slack.

Licensing & Compatibility

The project is released under the MIT License, which permits commercial use and modification with attribution.

Limitations & Caveats

This tool is strictly for research and learning assistance; it does not constitute investment advice. It avoids making predictions, issuing buy/sell recommendations, or guaranteeing returns. A core constraint is the absolute prohibition against fabricating information or misrepresenting original statements; any content not directly from the corpus is explicitly flagged as an inference based on his methodology.