A collection of investment research skills that applies the methodologies of four value investing masters via multi-agent adversarial analysis to generate structured, actionable investment reports. Users interact with the skills via an AI client such as Claude Code or Codex by providing a company name or stock ticker.
Project overview
The project structures AI-driven investment analysis by deploying four independent agents that research a company in parallel from different investing master perspectives, creating conflict and tension designed to avoid blind spots.
Project type
AI Agent · AI Coding
Use cases
Coding & Development
Deployment
Refer to project documentation
License
MIT
Best for
General users or researchers who want investment analysis structured around four distinct value investing master perspectives and delivered with actionable pass/fail/grey-area conclusions.
Key capabilities
Team-type skills launch four independent agents that research a company in parallel from different investing master perspectives.
Limitations and risks
The multi-agent adversarial analysis and deep research skills inherently consume a high volume of tokens, which affects users looking to manage or minimize model costs.
Operation requires external paid services, specifically AI clients like Claude Code or Codex. The data boundary requires external communication, and telemetry practices are not documented.
Getting started
Setup is rated as medium difficulty because it requires installing an AI client (such as Claude Code or Codex) and cloning the repository to copy the necessary skill files. The path to first success involves calling the skills in Claude Code, such as executing a command like /investment-research for a specific company.