investorskills  by xuboyuebobb

Portable investor judgment systems for AI finance agents

Created 2 months ago
343 stars

Top 80.3% on SourcePulse

GitHubView on GitHub
Project Summary

This open-source library provides a framework for structuring durable investing judgment into portable, machine-readable formats. It aims to make the thought processes of great investors accessible for human study and directly applicable by AI finance agents, enhancing financial analysis and decision-making.

How It Works

The core concept is "Investor Skills," which are plain-text skill packages designed for AI agents. Each skill typically consists of a SKILL.md file defining the agent's interface (when to use, inputs, process, outputs, guardrails) and an optional invest.md file detailing the structured investment judgment system. This system includes parameters like universe, market regime, signals, filters, sizing rules, and risk management, allowing for systematic application of complex investment strategies.

Quick Start & Requirements

  • Install: Use npx skills add https://github.com/xuboyuebobb/investorskills to install all skills, or npx skills add <URL> --skill "<skill-name>" for a specific skill.
  • Usage: Copy SKILL.md (and invest.md if present) into an AI agent's skill directory or directly into conversations with compatible agents (e.g., ChatGPT, Claude, Cursor).
  • Prerequisites: Requires Node.js and npm/yarn for npx.
  • Documentation: Links to docs/spec.md for schema details and templates are available within the repository.

Highlighted Details

  • Structured Judgment: The invest.md schema captures detailed investor logic, including signals, filters, sizing, risk parameters, market regime, and universe definitions, using YAML and Markdown.
  • Agent Interface: SKILL.md serves as the primary entry point for AI agents, specifying operational details and guardrails for skill execution.
  • Extensive Skill Library: The repository offers a wide array of pre-built skills based on prominent investors like Buffett, Livermore, Soros, and many others, covering diverse investment styles.
  • Curated Resources: Each skill is supplemented with relevant reading lists, podcasts, and YouTube resources to aid in understanding and stress-testing the investment logic.

Maintenance & Community

The project encourages contributions through provided templates for creating new investor skills and investment schema examples. Specific community channels (e.g., Discord, Slack) are not detailed in the README.

Licensing & Compatibility

  • License: MIT.
  • Compatibility: The permissive MIT license facilitates broad compatibility for commercial use and integration into closed-source applications.

Limitations & Caveats

The specification for the standalone INVEST.md format is explicitly marked as "alpha." While skills are designed for portability, they are noted as being "tailored for Questflow's financial harness," suggesting Questflow may provide the most integrated experience. The repository contains open-source skills, with exclusive, closed-source skills available to paid members within Questflow.

Health Check
Last Commit

1 week ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
343 stars in the last 30 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), Junyang Lin Junyang Lin(Core Maintainer at Alibaba Qwen), and
4 more.

ai-hedge-fund by virattt

0.1%
62k
AI-powered hedge fund proof-of-concept for educational use
Created 1 year ago
Updated 3 days ago
Feedback? Help us improve.