thesis-tracker

Investment Thesis Tracker: Post-Buy Discipline

Track an investment thesis for a holding. First use on a company → build the thesis; later uses → quarterly check.

Supported input: company name (auto-decide build vs track); company build (force rebuild); company quarterly check (against latest earnings).

"Buying is only the beginning. The real work is continuous tracking through the holding period." — Li Lu

"When the facts change, I change my mind. What do you do, sir?" — Keynes

Design Philosophy

Most investors' process is: research → buy → pray. Without systematic post-buy tracking, they: refuse to sell when they should ("wait, it'll come back"); panic-sell when they shouldn't ("down 20%, was I wrong?"); forget why they bought ("why did I buy this again?").

Buffett and Li Lu write down sell conditions before buying, then check the thesis each quarter.

Step 1: Decide the Mode

Check whether a thesis file already exists (use read_file on reports/{company}-thesis.md): not found → Build (Mode A); found → Track (Mode B).

---

Mode A: Build the Investment Thesis

A0: Data Collection

Use web_search to get the current price, valuation (PE/PB/dividend yield), and latest financials. If a research report exists, read_file it first. Use financial_rigor (command=verify_valuation) to verify valuation, financial_rigor (command=verify_market_cap) to sanity-check market cap.

A1: Core Thesis (must be writable in ≤200 words)

The thesis must answer 5 questions, one sentence each:

I bought {Company} at {price} because:
1. The essence of this business is ___, and I understand how it makes money.
2. Its moat is ___, and it's widening / stable.
3. Management is ___, trustworthy because ___.
4. The current price is a {discount}% to intrinsic value; the margin of safety comes from ___.
5. Even if I'm wrong, downside is contained because ___.

If you can't fill 5 sentences, the thesis itself is flawed — the buy decision isn't clear enough.