Market Sages brings 13 AI investor personas to your terminal
Developer hyhmrright rebuilt the ai-hedge-fund multi-agent stock analysis project as a single Claude Code skill file, letting users invoke 13 AI investor personas — from Buffett to Taleb — with a simple `/sages NVDA` command and no Python, API keys, or paid data subscriptions required.
Score breakdown
Developers and investors can explore multi-persona AI stock analysis workflows directly in Claude Code, Codex CLI, or Gemini CLI without any infrastructure setup, making it a practical reference for building prompt-only agentic skills that replace heavier orchestration stacks.
- 01Market Sages is a single `skill.md` file that works as a Claude Code skill, requiring no Python, API keys, or paid financial data.
- 02The `/sages` command invokes 13 AI investor personas: Buffett, Munger, Graham, Burry, Taleb, Cathie Wood, Druckenmiller, Ackman, Fisher, Lynch, Pabrai, Damodaran, and Jhunjhunwala.
- 03Each persona has its own signal rules, valuation lens, and voice style rather than a shared rubric.
Developer hyhmrright published Market Sages, a Claude Code skill that reimagines the `ai-hedge-fund` project as a zero-setup terminal tool. Where the original requires a Python environment, multiple LLM API keys, and a paid financial data API, Market Sages compresses the core value — investor personas and their analytical frameworks — into a single `skill.md` file. Invoking `/sages NVDA` triggers 13 AI impressions of legendary investors: Warren Buffett, Charlie Munger, Benjamin Graham, Michael Burry, Nassim Taleb, Cathie Wood, Stanley Druckenmiller, Bill Ackman, Philip Fisher, Peter Lynch, Mohnish Pabrai, Aswath Damodaran, and Rakesh Jhunjhunwala. Each sage applies its own signal rules and voice — Buffett discusses moats and owner earnings, Burry quotes EV/EBIT and FCF yield, Taleb declines price targets and lectures on tail risk — before a Portfolio Manager agent synthesizes a final action and conviction level.
Cross-CLI portability is built in via thin adapter files, so the same prompt runs in Codex CLI and Gemini CLI in addition to Claude Code.
The skill supports several usage patterns: users can paste in their own financial data for more grounded analysis, run multi-stock comparisons, filter to specific sages with `@handles`, or review a full portfolio allocation. Cross-CLI portability is built in via thin adapter files, so the same prompt runs in Codex CLI and Gemini CLI in addition to Claude Code. The post credits `@virattt` and `ai-hedge-fund` as the original inspiration and notes that Market Sages is explicitly an educational and entertainment tool — LLMs can hallucinate numbers, and the sages are AI impressions of investor thinking styles, not the actual investors. Docs are available in Chinese, Japanese, and Korean.
Key facts
- 01Market Sages is a single `skill.md` file that works as a Claude Code skill, requiring no Python, API keys, or paid financial data.
- 02The `/sages` command invokes 13 AI investor personas: Buffett, Munger, Graham, Burry, Taleb, Cathie Wood, Druckenmiller, Ackman, Fisher, Lynch, Pabrai, Damodaran, and Jhunjhunwala.
- 03Each persona has its own signal rules, valuation lens, and voice style rather than a shared rubric.
- 04A 'Portfolio Manager' synthesis agent delivers a final action and conviction level after all sages weigh in.
- 05Users can paste their own financial data directly into the prompt for more grounded analysis.
- 06Thin adapter files (`AGENTS.md`, `GEMINI.md`) extend compatibility to Codex CLI and Gemini CLI.