# Project Context

This is my ai agent workspace. I use it for research, content creation, and productivity workflows.

# About me

I create content about technology and productivity. My audience is people who want to learn about technology and want practical, no-nonsense tutorials. I prefer clear, jargon-free output.

# Rules

- Always ask clarifying questions before starting a complex task
- Show your plan and steps before executing
- Keep reports and summaries concise. Bullet points over paragraphs.
- Save all output files to the output folder.
- Cite sources when doing research. 

# Project Structure

- workflows/ - Workflow instruction files (plain english recipes the agent follows)
- output/ - finished deliverables (reports, drafts, analysis)
- resources/ - Reference docs and templates

i want to build a workflow where i can give you any stock, and you'll research it thoroughly, do the fundamental analysis, organize the findings, and produce a clean, structured report. At the top of the report, give a score out of 100. Before you start researching, you should ask me clarifying questions about the scope, the audience and what I specifically want to know.

Read c:\Users\rauns\first-agent\CLAUDE.MD

Agent:Explore project structure

I have a good picture of the project. It's a fresh workspace — only CLAUDE.MD exists, no folders yet. Let me ask a few questions before I write the plan.
User has answered your questions: 
- "What investment style should the analysis default to?"="All-purpose (Recommended)"
- "How should the 100-point score be structured?"="Single holistic score"
- "Should the report include an explicit Buy / Hold / Sell recommendation?"="Yes — include it (Recommended)"
- "Who will typically read these reports?"="Personal use"

You can now continue with the user's answers in mind.

Good, I have everything I need. Let me write the plan now.

Write C:\Users\rauns\.claude\plans\i-want-to-build-purrfect-milner.md
128 lines

Claude’s Plan C:\Users\rauns\.claude\plans\i-want-to-build-purrfect-milner.md
Stayed in plan mode

also include a recommendation for entry points for a position in the stock in the report. after tldr and before business overview, i want a one-liner on what the business does.

Edit C:\Users\rauns\.claude\plans\i-want-to-build-purrfect-milner.md
Added 2 lines
Done. Ready to proceed when you are.

Claude’s Plan C:\Users\rauns\.claude\plans\i-want-to-build-purrfect-milner.md
User approved the plan
Plan approved. Creating the three directories and two files now.

# Stock Analysis Workflow

## Purpose
Research any stock thoroughly, perform fundamental analysis, and produce a structured report
with a score out of 100, a Buy/Hold/Sell recommendation, and entry point guidance.
Reports are saved to `output/` and are written for personal use — dense, bullet-first, no fluff.

---

## Step 1 — Ask Clarifying Questions

Before doing any research, ask the user these four questions:

1. **Focus** — What specifically do you want to understand about this company?
   (e.g., growth story / financial health / competitive position / recent news / all of the above)

2. **Horizon** — What's your investment horizon for this analysis?
   (short-term trade / 1–3 years / long-term hold)

3. **Concerns** — Any specific concerns, claims, or rumors you want verified?

4. **Scope** — Anything to focus extra on, or skip entirely?

Do not begin research until these questions are answered.

---

## Step 2 — Research

Use WebSearch and WebFetch to gather data. Work through each area below in order.
**Cite every source with a URL** — no unsourced claims in the final report.

| Area | What to find |
|---|---|
| Business overview | Business model, products/services, revenue segments, company history |
| Recent financials | Revenue, EPS, operating margins, free cash flow — last 4 quarters |
| Balance sheet | Cash on hand, total debt, debt/equity ratio, interest coverage |
| Valuation | P/E, P/S, EV/EBITDA — current vs 5-year average and vs closest peers |
| Growth | Revenue and earnings growth rate, forward analyst estimates, guidance |
| Competitive moat | Market position, key competitors, differentiation, switching costs |
| Management | CEO track record, capital allocation history, insider ownership % |
| Recent news | Last 3–6 months: earnings beats/misses, guidance changes, M&A, macro impact |
| Analyst consensus | Buy/Hold/Sell count, average price target, current price, implied upside/downside |
| Risks | Regulatory, competitive, macro, balance sheet, and company-specific risks |
| Entry points | Current price, 52-week range, 50-day and 200-day moving averages, key support/resistance levels, upcoming catalysts (earnings dates, product launches, regulatory decisions) |

---

## Step 3 — Score and Recommendation

After completing research, assign:

**Score (out of 100)** — A single holistic number reflecting overall investment quality.
Write 1–2 sentences justifying the score. Do not show the internal rubric in the report.

Internal scoring guidance (do not include in report):
- **80–100** Exceptional — strong business, healthy financials, attractive valuation, clear catalysts
- **60–79** Solid — good business with moderate risk or a valuation concern
- **40–59** Mixed — meaningful risks or weak financials offset real positives
- **Below 40** Avoid — poor fundamentals, stretched valuation, or serious red flags

**Recommendation** — State one of: **BUY / HOLD / SELL** with a one-line rationale.

---

## Step 4 — Write the Report

Fill in the template at `resources/stock-report-template.md`.

Save the completed report to:

output/[TICKER]-analysis-[YYYY-MM-DD].md


Confirm the save path with the user when done.

---

## Quality Checklist

Before saving, verify:
- [ ] All 11 report sections are populated (no placeholders remaining)
- [ ] Score and recommendation appear at the very top
- [ ] "What they do" one-liner is present after TL;DR
- [ ] Entry points section includes current price, zones, and suggested approach
- [ ] Every factual claim has a source URL in the Sources section
- [ ] Filename follows the `[TICKER]-analysis-[YYYY-MM-DD].md` format

run the stock analysis workflow on Nvidia

# NVIDIA Corporation (NVDA) — Stock Analysis
Date: 2026-04-30

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
SCORE: 78/100  |  BUY
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
World-class AI compute monopoly with near-irreplaceable software moat; forward P/E of ~24x sits well below its own 5-year historical average, making valuation far more defensible than the headline trailing multiple implies. Primary risks — China export headwind (~$8B/quarter in lost revenue) and gradual custom-silicon displacement at inference workloads — are real but manageable on a 1–3 year horizon.

---

## TL;DR
- FY2026 revenue hit $215.9B (+65% YoY); FY2027 analyst consensus is $369B (+71%) driven by Blackwell ramp and Vera Rubin introduction.
- Data center is 90% of revenue ($193.7B in FY2026); hyperscaler capex still accelerating.
- China export controls permanently removed ~$8B/quarter in addressable revenue — this is now priced in, not a new surprise.
- Forward P/E of ~24x is below the 5-year average of 68x; the stock is cheaper on a growth-adjusted basis than it has been in years.
- Custom silicon (Google TPU, Amazon Trainium, Microsoft Maia) is a real long-term threat to inference share, but CUDA's ecosystem lock-in keeps training spend anchored at Nvidia for the foreseeable future.

**What they do:** Nvidia designs and sells GPUs and accelerated computing platforms — primarily to hyperscalers and enterprises for AI model training and inference — with its proprietary CUDA software stack creating deep switching costs that make it the default infrastructure layer for AI development.