# xbtlin/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.

Repository: https://github.com/xbtlin/ai-berkshire
Canonical: https://ross.abutalabs.com/products/ai-berkshire
Homepage: https://github.com/xbtlin/ai-berkshire#readme
Language: Python
License: MIT
License Family: permissive
Topics: ai, ai-agent, anthropic, berkshire-hathaway, charlie-munger, china-stock, claude, claude-code, financial-analysis, fintech, fundamental-analysis, investment, investment-research, llm, mcp, portfolio-management, stock-analysis, stock-market, value-investing, warren-buffett
Last push: 2026-08-26T16:52:51+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 46, longevity 10
- inputs: {"age_days": 148, "days_push": 7, "days_rel": 148, "gap_med": null, "n_releases_24m": 1}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 15907, forks 2378 (observed 2026-08-28T04:11:14.408852+00:00)

## What it is
A collection of AI research skills for Claude Code and Codex that systematizes the value investing methodologies of Buffett, Munger, Duan Yongping, and Li Lu into structured multi-agent analysis workflows. It produces forced-conclusion investment reports with adversarial multi-perspective scoring, anti-bias checks, and financial calculation verification tools.

## Use cases
- analyze whether a stock is worth buying with AI
- run value investing research on a company
- get structured buy/pass verdicts with price ranges instead of hedged AI answers
- apply Buffett Munger Duan Yongping Li Lu methodologies to stock analysis
- verify market cap and financial metrics against LLM hallucinations
- screen companies using value investing criteria
- run multi-agent adversarial investment analysis

## When to choose
- you use Claude Code or Codex and want disciplined, structured value investing research
- you want multi-perspective analysis with forced conclusions rather than generic AI answers
- you need built-in anti-bias mechanisms like inversion checks and quick-reject lists
- you want reproducible financial calculation verification in Python

## When to avoid
- you need real-time market data feeds or automated trading execution
- you expect guaranteed returns or professional investment advice
- you don't use Claude Code or Codex as your AI coding environment
- you need quantitative/backtesting frameworks rather than qualitative research workflows

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, prompt-engineering, mcp, data-science, analytics
- domain: fintech, artificial-intelligence, large-language-models, analytics
- platform: cli, python, cross-platform
- tags: value-investing, claude-code, codex, stock-analysis, financial-analysis, multi-agent, warren-buffett, charlie-munger, investment-research, skills-collection, ai-agents

## Member repositories
- xbtlin/ai-berkshire (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:14.408852+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:05:19.939534+00:00, confidence not recorded.
  - readme: https://github.com/xbtlin/ai-berkshire (fetched 2026-08-28T04:11:14.408852+00:00, sha 6dfcee3d3186)
  - homepage: https://github.com/xbtlin/ai-berkshire#readme (fetched 2026-08-29T08:03:12.642649+00:00, sha af555f0a3bde)
- Data as of 2026-08-30T08:39:29.467469+00:00.