ai-berkshire
AI 时代的伯克希尔：基于 Claude Code / Codex 的价值投资研究框架。巴菲特·芒格·段永平·李录四大师方法论 + 多Agent并行研究。| AI-era Berkshire: a value investing research framework built for Claude Code / Codex. 4 masters' methodologies + multi-agent adversarial analysis.
- stars
- 16,417
- 30-day movement
- +6622/day
- Related entries
- 60
- Connections
- 3
ai-berkshire is a value investing research framework built for Claude Code / Codex, described as encoding the methodologies of Buffett, Munger, Duan Yongping, and Li Lu alongside multi-agent parallel/adversarial analysis. It ships as an MCP server with AGENTS.md and CLAUDE.md files, and its topics indicate stock, fundamental, and portfolio analysis including China A-shares.
Use it if you already work in Claude Code or Codex and want a structured, methodology-driven framework for value investing research rather than building your own agent setup.
Use it to
- Run multi-agent adversarial analysis on a stock
- Apply the four masters' value investing methodologies
- Set up investment research agents via AGENTS.md / CLAUDE.md
- Analyze fundamentals and China stocks through Claude Code
- Research portfolios with an MCP server workflow
For Value investors and analysts already using Claude Code or Codex.
- Role
- mcp-server
- Language
- HTML
- Licence
- MIT
- Forks
- 2,447
- Open issues
- 24
- Last push
- 2026-09-11
- Latest release
- v1.0.0 · 2026-04-07