# How the Buffett Master Perspective Shapes Investment Decisions in AI-Berkshire

> Discover how the Buffett master perspective guides AI-Berkshire investments. Learn about financial valuation, moat assessment, and a six-gate checklist for value-driven decisions.

- Repository: [Xbt Lin/ai-berkshire](https://github.com/xbtlin/ai-berkshire)
- Tags: deep-dive
- Published: 2026-07-29

---

**The Buffett master perspective applies a disciplined three-layer filter—financial valuation, moat assessment, and a six-gate pre-buy checklist—to enforce value-driven investing only "at a price well below intrinsic value."**

The AI-Berkshire repository codifies Warren Buffett’s investment philosophy into executable Codex skills. This framework implements the Buffett master perspective through rigorous cash-flow-based multiples, competitive advantage scoring, and a rapid 10-minute screening process designed to reject any idea failing critical guardrails from the outset.

## Core Components of the Buffett Master Perspective

The Buffett master perspective in AI-Berkshire is structured as three tightly-coupled validation layers. According to the repository’s [`README_EN.md`](https://github.com/xbtlin/ai-berkshire/blob/main/README_EN.md) (lines 7–9), each layer enforces a specific Buffett principle: price versus value, durable competitive advantage, and pre-investment discipline.

### Financial Valuation: Price Versus Value

This layer implements Buffett’s maxim that *“price is what you pay, value is what you get.”* The [`codex-prompts/valuation.md`](https://github.com/xbtlin/ai-berkshire/blob/main/codex-prompts/valuation.md) skill (generated to [`skills/valuation.md`](https://github.com/xbtlin/ai-berkshire/blob/main/skills/valuation.md)) computes **Ex-cash P/E** by stripping out cash and non-core assets to derive a “pure” earnings multiple.

The valuation logic in [`tools/financial_rigor.py`](https://github.com/xbtlin/ai-berkshire/blob/main/tools/financial_rigor.py) benchmarks the current P/E against a 10-year historical median. A significant discount triggers a **Buffett-style “genuinely cheap”** signal. The framework explicitly calculates the **margin of safety** as the gap between current price and intrinsic value, rejecting candidates where this buffer is insufficient.

### Moat Assessment: Durable Competitive Advantage

Referencing [`README_EN.md`](https://github.com/xbtlin/ai-berkshire/blob/main/README_EN.md) lines 394–395, this layer evaluates whether a business possesses a protective moat. The **three-layer moat model** scores:

- **Network Effects** – Customers become more valuable as the network grows.
- **Switching Costs** – High cost for customers to change vendors.
- **Scale Economies** – Cost advantages that grow with production volume.

Each layer receives a star rating (★ to ★★★★★). Only companies scoring **≥ ★★★** overall earn a “Buffett-approved moat,” allowing progression to deeper valuation analysis.

### The Six-Gate Pre-Buy Checklist

As documented in [`README_EN.md`](https://github.com/xbtlin/ai-berkshire/blob/main/README_EN.md) lines 436–440, the [`skills/investment-checklist.md`](https://github.com/xbtlin/ai-berkshire/blob/main/skills/investment-checklist.md) skill enforces Buffett’s rule: *“If you can’t answer these six questions, walk away.”* This 10-minute filter forces analysts to verify:

1. **Business Understandability** – Do you grasp the key drivers?
2. **Management Quality** – Is capital allocated wisely?
3. **Economic Moat** – Is the moat wide and widening?
4. **Financial Health** – Is the balance sheet strong?
5. **Valuation Discipline** – Is the price “genuinely cheap”?
6. **Long-Term Outlook** – Can you see the business 10 years ahead?

If any gate fails, the analysis stops immediately.

## Architectural Implementation and Command Usage

The Buffett master perspective is encoded as **Codex skills** invoked via command line or Claude-compatible slash-commands. The pipeline, visualized in `assets/architecture-en.svg`, flows from **Data Collection → Business Essence (Duan) → Moat (Buffett) → Inversion (Munger) → Management (Duan + Buffett) → Civilizational Trends (Li Lu)**. The Buffett layer sits between “Business Essence” and “Management Assessment,” ensuring only moat-qualified companies reach valuation.

Execute the Buffett workflow using these commands:

```bash

# Run the six-gate pre-buy checklist on a ticker (e.g., AAPL)

$ python3 -m codex.run /investment-checklist AAPL

# Returns pass/fail status for each of the six gates.

```

```bash

# If the checklist passes, invoke Ex-cash P/E valuation

$ python3 -m codex.run /valuation AAPL --method=excash_pe

# Outputs intrinsic value, margin of safety, and star rating.

```

```bash

# Assess the moat rating independently

$ python3 -m codex.run /moat-rating AAPL

# Returns a ★★★★ score; combine with checklist results for final decision.

```

You can script a final decision rule:

```bash
if checklist_passed and moat_score >= 3 and margin_of_safety > 15%:
    echo "Buy"
else
    echo "Hold/Skip"

```

After editing any skill in `skills/*.md` or `codex-prompts/`, run `python3 scripts/sync-codex-skills.py` to regenerate the executable Codex files.

## Key Source Files

- **[`README_EN.md`](https://github.com/xbtlin/ai-berkshire/blob/main/README_EN.md)** – Core documentation defining Buffett’s valuation principles (lines 7–9), moat definitions (lines 394–395), and the six-gate checklist (lines 436–440).
- **[`skills/investment-checklist.md`](https://github.com/xbtlin/ai-berkshire/blob/main/skills/investment-checklist.md)** – Implements the executable six-gate Buffett pre-buy guardrails.
- **[`codex-prompts/valuation.md`](https://github.com/xbtlin/ai-berkshire/blob/main/codex-prompts/valuation.md)** – Source for [`skills/valuation.md`](https://github.com/xbtlin/ai-berkshire/blob/main/skills/valuation.md); contains Ex-cash P/E logic and margin of safety calculations.
- **[`tools/financial_rigor.py`](https://github.com/xbtlin/ai-berkshire/blob/main/tools/financial_rigor.py)** – Helper utilities for computing historical P/E medians and safety margins.
- **`assets/architecture-en.svg`** – Pipeline diagram showing the Buffett layer positioned between Business Essence and Management Assessment.

## Summary

- The **Buffett master perspective** enforces a three-layer validation: financial valuation (price vs. value), moat assessment (durable advantage), and a six-gate pre-buy checklist.
- **Ex-cash P/E** and **margin of safety** calculations in [`tools/financial_rigor.py`](https://github.com/xbtlin/ai-berkshire/blob/main/tools/financial_rigor.py) operationalize Buffett’s value investing metrics.
- The **three-layer moat model** (Network Effects, Switching Costs, Scale Economies) requires a minimum ★★★ rating to proceed.
- All components are exposed as **Codex skills** (`/investment-checklist`, `/valuation`, `/moat-rating`) within the AI-Berkshire pipeline.
- The architecture ensures that only companies passing the Buffett filter advance to deeper management and trend analysis.

## Frequently Asked Questions

### What is the Buffett master perspective in AI-Berkshire?

The Buffett master perspective is the repository’s implementation of Warren Buffett’s investment philosophy, encoded as executable Codex skills. It combines rigorous valuation metrics, competitive moat scoring, and a disciplined six-question checklist to filter investment opportunities before deep analysis begins.

### How does the six-gate pre-buy checklist work?

The checklist, defined in [`skills/investment-checklist.md`](https://github.com/xbtlin/ai-berkshire/blob/main/skills/investment-checklist.md) and referenced in [`README_EN.md`](https://github.com/xbtlin/ai-berkshire/blob/main/README_EN.md) lines 436–440, forces analysts to answer six critical questions about business understandability, management quality, moat width, financial health, valuation discipline, and long-term outlook. If any gate fails, the investment is rejected immediately, mirroring Buffett’s “stop-when-you-don’t-know” rule.

### What constitutes a "Buffett-approved" moat?

A Buffett-approved moat requires a minimum three-star rating (★★★) across three assessed dimensions: network effects, switching costs, and scale economies. This rating system, documented in [`README_EN.md`](https://github.com/xbtlin/ai-berkshire/blob/main/README_EN.md) lines 394–395, ensures only businesses with durable competitive advantages proceed to valuation calculations.

### How is intrinsic value calculated in the Buffett model?

Intrinsic value is derived using the **Ex-cash P/E** method, which strips cash and non-core assets from the valuation, then compares the resulting multiple against a 10-year historical median. The **margin of safety** represents the percentage difference between this intrinsic value and the current market price, with wider gaps signaling stronger Buffett-style buying opportunities.