Look Through Earnings

Name: Look Through Earnings
Author: kangarooking

kangarooking/buffett-letters-skill

45 installs
65 repo stars
Updated April 17, 2026
kangarooking/buffett-letters-skill

Evaluate business value by attributing portion of retained earnings to ownership.

About

Reported earnings alone mislead; attribute a share of retained earnings reinvested in the business to shareholders. True owner earnings = reported + retained attributed share.

Owner earnings calculation
Earnings quality assessment

Look Through Earnings by the numbers

45 all-time installs (skills.sh)
+1 installs in the week ending Aug 2, 2026 (Skillselion tracking)
Ranked #606 of 1,106 Finance & Trading skills by installs in the Skillselion catalog
Data as of Aug 5, 2026 (Skillselion catalog sync)

npx skills add https://github.com/kangarooking/buffett-letters-skill --skill look-through-earnings

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

[![Listed on Skillselion](https://skillselion.com/badge/skills/kangarooking/buffett-letters-skill/look-through-earnings.svg)](https://skillselion.com/skills/kangarooking/buffett-letters-skill/look-through-earnings)

Installs	45
repo stars	★ 65
Last updated	April 17, 2026
Repository	kangarooking/buffett-letters-skill ↗

What it does

Evaluate business value by attributing portion of retained earnings to ownership.

Files

SKILL.mdMarkdownGitHub ↗

透视盈余（含所有者收益/反EBITDA）

R — 原文 (Reading)

"We feel that the undistributed earnings of companies we hold in large amounts are fully as important to us as those we report." (1978)

"References to EBITDA make us shudder — does management think the tooth fairy pays for capital expenditures?" (2000)

I — 方法论骨架

两个核心概念合并为一个skill：

透视盈余: 真实经济收益 = 报告利润 + 被投资公司未分配利润份额 - 假设税款

会计只显示已收股息，但未分配利润也是你的（如果有控制权或长期持有）

所有者收益: 净利润 + 折旧摊销 - 维持性资本支出 ≠ EBITDA

折旧不是"非现金费用"——它是预付资本的回收
EBITDA忽略了维持竞争地位所需的真实资本支出

A1 — 书中的应用

案例: Apple的透视盈余

问题: 伯克希尔从Apple投资中获得多少？
使用: 2021年伯克希尔享有的Apple收益=$56亿，但仅收股息=$7.85亿
结果: 会计利润严重低估真实经济进展（差7倍）

A2 — 触发场景

1. 分析一家利润高但资本支出也高的公司 2. 评估一家控股/参股公司的真实盈利能力 3. 有人用EBITDA估值——需要指出问题 4. 理解为什么净利润和现金流差距很大

语言信号

"这家公司EBITDA很高"
"折旧不是现金成本"
"利润和现金流为什么差这么多？"
"未分配利润算不算我的？"

与相邻 skill 的区分

与 economic-moat 的区别：护城河分析企业竞争力，透视盈余分析企业真实赚钱能力

E — 可执行步骤

1. 计算所有者收益 = 净利润 + 折旧摊销 - 维持性资本支出

完成标准：区分"维持性"vs"扩张性"资本支出

2. 检查EBITDA陷阱

完成标准：如果EBITDA >> 所有者收益，标注"盈利质量存疑"
判停：如果EBITDA与自由现金流差距>50% → 发出警告

3. 计算透视盈余（适用于有参股公司的情况）

完成标准：报告利润 + 按持股比例的未分配利润 - 假设税款

B — 边界

估计"维持性"vs"扩张性"资本支出需要判断——不同人可能得出不同结论
透视盈余对短期投资者意义不大（未分配利润可能永远不实现）
作者盲点：巴菲特有足够的控制权来确保未分配利润最终实现——散户没有

---

审计信息

验证通过: V1 ✓ / V2 ✓ / V3 ✓
蒸馏时间: 2026-04-16

{
  "skill": "look-through-earnings",
  "version": "0.1.0",
  "source_book": "《巴菲特致股东的信》 — 沃伦·巴菲特",
  "darwin_compatible": true,
  "test_cases": [
    {
      "id": "should-trigger-01",
      "type": "should_trigger",
      "prompt": "这家公司EBITDA很高，是不是很赚钱",
      "expected_behavior": "应激活 look-through-earnings，指出EBITDA忽略了资本支出这个真实成本",
      "notes": "EBITDA误导的典型场景"
    },
    {
      "id": "should-trigger-02",
      "type": "should_trigger",
      "prompt": "净利润和自由现金流为什么差这么多",
      "expected_behavior": "应激活 look-through-earnings，解释所有者收益=净利润+折旧-维持性资本支出",
      "notes": "盈利质量评估"
    },
    {
      "id": "should-trigger-03",
      "type": "should_trigger",
      "prompt": "折旧是非现金费用，应该加回来对吗",
      "expected_behavior": "应激活 look-through-earnings，指出折旧是真实的资金预付，牙仙不会为资本支出买单",
      "notes": "折旧认知误区"
    },
    {
      "id": "should-not-trigger-01",
      "type": "should_not_trigger",
      "prompt": "这个公司有没有护城河",
      "expected_behavior": "不应激活本skill，这是竞争分析",
      "notes": "诱饵：竞争优势≠盈余质量"
    },
    {
      "id": "should-not-trigger-02",
      "type": "should_not_trigger",
      "prompt": "什么时候该卖出",
      "expected_behavior": "不应激活本skill，这是卖出时机决策",
      "notes": "诱饵：卖出决策≠盈余分析"
    },
    {
      "id": "edge-01",
      "type": "edge_case",
      "prompt": "SaaS公司用订阅制，收入确认和传统不一样",
      "expected_behavior": "应指出会计差异不等于盈余质量问题——需要区分会计方法和真实经济差异",
      "notes": "边界：会计方法差异≠盈余操纵"
    }
  ],
  "minimum_pass_rate": 0.8
}