Skip to content
Sign in
  • Browse all toolsFilter by category and pricing
  • Browse categoriesWriting, code, image, audio and more
  • Compare toolsPut up to three side by side
  • Read the blogGuides and comparisons

AI Berkshire

Open-source value investing research skills for Claude Code and Codex.

AI Berkshire is a free, open-source value investing research framework for Claude Code and Codex. AI Berkshire combines structured company analysis, parallel research agents, financial verification and portfolio workflows for investors and investment researchers. The MIT-licensed repository provides installable skills, while deep research can incur model usage costs.

Updated 9 Nov 20266 sections

At a glance

Pricing
FreeThe project charges no fees and uses the MIT license. Deep research workflows have high token consumption; model usage costs remain separate.
Added
9 Nov 2026
AI Berkshire website screenshot

What is AI Berkshire?

AI Berkshire is an open-source value investing research framework built for Claude Code and Codex. AI Berkshire translates the methods of Warren Buffett, Charlie Munger, Duan Yongping and Li Lu into structured research skills. The framework supports investors examining business quality, competitive advantages, management, valuation and investment risks.

AI Berkshire organizes research through skills, agents and supporting tools. Individual skills provide entry points for specific tasks, while team-based skills coordinate independent research perspectives. The repository also contains completed research reports, and a separate reading site organizes reports by company, topic and time with search and reading-progress support.

What AI Berkshire can do

AI Berkshire covers company research, earnings analysis, industry screening, portfolio management and investment thinking tools. AI Berkshire uses structured conclusions, information-quality ratings and uncertainty labels rather than treating every finding as equally reliable. Valuation workflows include scenario analysis and reverse valuation, while financial tools check calculations and compare important data across independent sources.

  • AI Berkshire researches listed companies, private companies and management teams.
  • AI Berkshire reviews original financial statements rather than relying on secondary research reports.
  • AI Berkshire screens industries, indexes and themes for companies worth further study.
  • AI Berkshire evaluates portfolio concentration, position sizing and rebalancing.
  • AI Berkshire tracks investment theses and distinguishes factual, valuation and wording changes.
  • AI Berkshire verifies market capitalization and uses precise decimal calculations.

How AI Berkshire works

AI Berkshire installs into an existing Claude Code or Codex environment after the user clones the GitHub repository. The project supplies installation scripts for macOS, Linux and Windows. Claude Code users invoke research commands directly, while Codex users restart the client after installation and request tasks by skill name.

AI Berkshire team workflows assign independent agents to research the same company, search sources, cross-check data and reach their own conclusions. A team lead then synthesizes those findings. Lightweight skills bypass that team layer for narrower tasks. Optional Codex slash prompts provide an additional command-style entry point alongside the installed skills.

Who AI Berkshire is for

AI Berkshire serves value investors and researchers who want repeatable company analysis within Claude Code or Codex. AI Berkshire supports both initial screening and deeper investigation, including management assessments and research into information-scarce private companies. Standardized structures also help compare companies, revisit earlier conclusions and align research produced by different team members.

  • AI Berkshire helps investors screen companies before committing to deeper research.
  • AI Berkshire supports researchers analyzing business quality, valuation and management.
  • AI Berkshire helps portfolio holders review allocations and monitor investment theses.
  • AI Berkshire supports analysts investigating industry opportunities and supply-chain constraints.

AI Berkshire pricing and access

AI Berkshire is available without a project fee through its public GitHub repository under the MIT license. The source explicitly states that the project does not collect payments, recommend stocks through paid groups or accept money in any form. Access to the framework is distinct from the resources needed to run research through a compatible AI client.

AI Berkshire warns that deep research skills use substantial tokens because they perform repeated research, cross-validation and multi-agent synthesis. To control costs, the documentation recommends starting with narrower workflows such as quality screening or news attribution, then escalating to comprehensive company research only when the initial findings justify further investigation.

AI Berkshire strengths and limitations

AI Berkshire combines consistent report structures with adversarial research perspectives and explicit data checks. These mechanisms support comparable research outputs and make uncertainty visible. AI Berkshire also allows insufficient evidence to remain a gray area rather than forcing certainty. The source cautions that model capability matters for complex judgments involving competitive advantages, valuation, management and risk.

  • AI Berkshire requires client installation and skill setup rather than offering only a browser-based workflow.
  • AI Berkshire deep research can carry substantial token costs.
  • AI Berkshire research quality remains dependent on the chosen model.

AI Berkshire FAQ

What is AI Berkshire?

AI Berkshire is an open-source value investing research framework for Claude Code and Codex. AI Berkshire structures the methods of Warren Buffett, Charlie Munger, Duan Yongping and Li Lu into skills for company research, financial analysis, industry screening and portfolio management.

Is AI Berkshire free?

AI Berkshire is available without a project fee under the MIT license through its public GitHub repository. AI Berkshire does not collect payments, but deep research workflows consume model tokens, so running the framework can involve separate model usage costs.

Does AI Berkshire work with Claude Code and Codex?

AI Berkshire supports both Claude Code commands and Codex skills. AI Berkshire provides installation scripts for macOS, Linux and Windows, plus optional Codex slash prompts. Users clone the repository, run the relevant installation script and invoke research tasks within their chosen client.

What can you use AI Berkshire for?

AI Berkshire supports company valuation, management research, financial statement review, industry screening and portfolio analysis. AI Berkshire also provides workflows for private-company research, stock-price news attribution, investment thesis tracking and comparisons that distinguish factual changes from valuation or wording changes.

How does AI Berkshire check financial data?

AI Berkshire cross-checks key financial data against independent sources and provides market capitalization verification tools. AI Berkshire uses Python decimal.Decimal for precise calculations, assigns information-quality ratings and marks insufficient evidence as uncertain rather than presenting unsupported estimates as established facts.

Does AI Berkshire use multiple agents?

AI Berkshire uses parallel independent agents in team-based investment and earnings research workflows. AI Berkshire assigns separate research perspectives to search, analyze and reach conclusions before a team lead synthesizes the results. Lightweight skills instead connect directly to supporting tools without that team layer.

From the community

No reviews yet

Be the first to review this tool. What you write helps the next person choose.

More tools like AI Berkshire

Candlestick

www.candlestick.ai

AI-powered stock investing with customizable portfolios, automatic investing, and conversational financial explanations.

Paid

daily_stock_analysis

www.github.com

Open-source AI stock analysis with multi-market data, decision reports, strategy chat, and automated notifications.

Free

FinRobot

www.github.com

Open-source financial AI agents for equity research, code-computed valuations, and traceable investment reports.

Free