Learn

Use Cases

End-to-end journeys by investor type

Open Raw

Use Cases & Journeys

The Cookbook shows individual skills and their output. This page shows people with goals chaining skills end-to-end — including a worked case where the analysis changed the conclusion, and the common ways analyses go wrong. Commands use Claude Code slash syntax; on other platforms, reference the matching prompts/*.md file instead.

Journeys: Dividend investor · First-time investor · Earnings-season trader · Macro allocator · The skeptic · Worked case study · Anti-patterns


The Dividend / Income Investor

Goal: build a portfolio that pays reliable, growing income — without walking into a yield trap.

/dividend-analysis JNJ      ← safety score, payout ratio, coverage, growth streak
/dividend-analysis KO
/dividend-analysis O
/portfolio-review           ← paste candidates; check concentration & blended yield

What to look for: payout ratio under ~60%, coverage > 2x, positive FCF, and a multi-year growth streak. A 9% yield with a 110% payout ratio is a warning, not a bargain — the market is pricing in a cut.

Decision rule: target a blended portfolio yield of ~3%+ with an average safety score above 7.5, and no single holding over ~10%.


The First-Time Investor

Goal: “I have some cash and want to evaluate my first stock without pretending I’m a pro.”

/stock-eval AAPL            ← one screen: quality, value, moat, risk

Then don’t just read the score — decode it. Hit an unfamiliar term (Piotroski F-Score? ROIC?) → open the Glossary. Want the intuition → Concepts. Then sanity-check before trusting it:

/result-validator           ← paste the stock-eval output; get a confidence score

Mindset: the signal block is a starting point, not a verdict. A “BUY / MODERATE” with LOW confidence means “interesting, but the data is thin — learn more before acting.” Start small, size positions you can afford to be wrong on, and read margin of safety.


The Earnings-Season Trader

Goal: have a plan before the call, react with discipline after it.

# Before earnings
/fundamental-analysis NVDA          ← baseline: what does "good" look like this quarter?
/options-analysis NVDA --earnings   ← what move is the market pricing in (IV)?

# After the call
/earnings-call-analysis NVDA        ← paste transcript: tone, guidance delta, hidden risks
/technical-analysis NVDA            ← did price confirm or fade the reaction?

What to look for: the gap between guidance and expectations, language shifts versus prior quarters, and whether realized volatility justified the implied volatility you paid for. The edge is in the delta, not the headline number.


The Macro / Top-Down Allocator

Goal: position by reading the macro weather, then drill down to a name.

/economics-analysis                 ← regime: yield curve, Fed stance, leading indicators
/sector-analysis                    ← which sectors lead in this regime? relative strength
/stock-eval JPM                     ← best-in-class name within the favored sector
/portfolio-review                   ← does adding it reduce or add rate sensitivity?

What to look for: the macro regime sets the tilt — rising rates favor financials and value; easing favors growth and rate-sensitive yield. Pick the sector then the stock, and check the portfolio-level effect before you add risk.


The Skeptic / Risk Manager

Goal: actively hunt for reasons not to buy — the most underrated discipline in investing.

/financial-report-analyst TSLA 10-K   ← accounting red flags, one-offs, footnote risks
/insider-trading TSLA                  ← are insiders selling into strength?
/short-interest TSLA                   ← is smart money positioned against it?
/result-validator                      ← does the bull case survive scrutiny?

What to look for: disagreement between signals. Bullish fundamentals + heavy insider selling + rising short interest is a contradiction worth resolving before you commit. The goal isn’t to be negative — it’s to make the bull case earn it.


Worked Case Study: When Analysis Changes the Answer

A short, illustrative walkthrough (hypothetical ticker “XYZ”) showing how a thesis should evolve as evidence accumulates — not lock in at step one.

Step Skill Finding Running signal
1 stock-eval High ROIC, clean Piotroski 8/9, reasonable P/E BULLISH · MED conf
2 stock-valuation DCF + comps say ~15% undervalued BULLISH · stronger
3 financial-report-analyst 10-K reveals revenue increasingly from one customer + a large receivable build Confidence drops
4 insider-trading Two officers sold large blocks last month Signal weakens to NEUTRAL
5 result-validator Flags the customer-concentration + insider conflict; confidence LOW HOLD, not BUY

Lesson: steps 1–2 looked like a clean buy. The judgment came from steps 3–5, where the filing and insider data introduced risks the headline metrics hid. The final call — HOLD pending the next 10-Q — is the output of the disagreement, exactly what a composite signal block is designed to surface. A high score with low confidence is a “look closer,” not a “go.”


Anti-Patterns: How Analyses Go Wrong

Anti-pattern Why it bites The fix
Trusting a single DCF point estimate Tiny WACC/terminal-growth changes swing it wildly Read the bear/base/bull range and sensitivity table, not the single number
Ignoring the Confidence field A bullish score with LOW confidence is a guess Treat low confidence as “gather more data,” not “go”
Using stale or guessed data The AI reasons over whatever you give it Paste current, primary-source financials; see Data & Accuracy
Mistaking a yield trap for income A 10% yield often signals a coming cut Check payout ratio and FCF, not just yield
Anchoring on one signal block One skill sees one slice Run research-bundle / cross-check with result-validator
Confusing confidence with conviction They’re different axes See the signal-block anatomy
Skipping validation on composites Disagreeing sub-signals get averaged away Always result-validator a full-report / research-bundle

Next: Choose a Skill to build your own chain · Concepts for the reasoning · Data & Accuracy before you act.

Educational content only. Not financial advice. Tickers used illustratively, not as recommendations.