investment-memo-craft

Structures investment research into decision-ready Markdown research reports with valuation and action guidance.

16.4k|2.5k|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/xbtlin/ai-berkshire --skill investment-memo-craft-xbtlin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investment-memo-craft
Source: https://github.com/xbtlin/ai-berkshire/tree/main/codex-skills/investment-memo-craft
Command: npx skills add https://github.com/xbtlin/ai-berkshire --skill investment-memo-craft-xbtlin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Raw investment research often reads as balanced but indecisive analysis that investors cannot act on. This Skill converts company, industry, or fund research into a structured Codex research report with concrete business mechanics, falsifiable moat analysis, inverse thinking, scenario valuation, and explicit buy/hold/sell price bands. ## Core Features & Use Cases - Structured Report Workflow: Enforces a nine-step report shape covering core data, business essence, moat scoring, bear cases, management capital allocation, industry trends, valuation with safety margin, and a final decision memo. - Decision-Ready Output: Separates "good business" from "good investment at this price", provides distinct advice for empty-handed investors versus holders, and defines add/trim/sell triggers. - Typography Standards: Applies calm GitHub Markdown layout rules including dated filenames, metadata blocks, Chinese step headings, restrained bolding, and signed growth figures. - Use Case: After gathering financial data on a company like Pinduoduo, use this Skill to rewrite the findings into a long-form research report with moat scores, a short-seller-style bear case, reverse-DCF valuation bands, and an action table for different investor types. ## Quick Start Use investment-memo-craft to turn my research notes on this company into a polished Codex investment research report with valuation bands and an action checklist.

Frequently Asked Questions about investment-memo-craft

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write an investment research report with AI?

Provide your company research data and ask for a research report. The Skill structures it into nine steps from core data overview through moat assessment, bear-case analysis, valuation with price bands, and a final decision memo with actions for different investor types.

What sections should a value investing research report include?

This Skill mandates sections for core financial data, business essence in one sentence, moat scoring by source, failure paths with probabilities, management capital allocation history, industry value capture, scenario valuation with safety margin, and a closing action checklist.

Can I use this Skill with Claude Code?

No, this is a Codex-only skill kept under codex-skills/ for simple installation. The documentation explicitly states not to create a same-named Claude Code slash-command source unless intentionally adopting the workflow there.

Does this Skill fetch financial data or do valuation calculations?

No, it is a writing and layout overlay only. It pairs with the financial-data skill for source validation, investment-research for the Buffett/Munger framework, and report_audit.py before a report is treated as publishable.

When should the output be called a memo versus a research report?

Long-form outputs are titled research reports by default, with the full decision table placed near the end after the evidence. The investment memo framing is used only when the user explicitly requests a memo format.