AI-Powered Accounting Analysis Capstone

Follow these steps to connect your AI tool to SEC EDGAR using Python

Step 1: Check Python Version (3.11+ required)
python --version
# or on some systems:
python3 --version

Step 2: Install SEC EDGAR MCP

Option A - Using pip (recommended):
pip install sec-edgar-mcp

Option B - Using uvx (no installation, runs directly):
pip install uvx
# Then use uvx in your config (see Step 3)

Step 3: Configure Your AI Client

For Claude Desktop:
Edit config file at:
- Mac/Linux: ~/.config/claude-desktop/config.json  
- Windows: %APPDATA%\Claude\claude-desktop\config.json

Add this configuration:

If you used pip install:
{
  "mcpServers": {
    "sec-edgar-mcp": {
      "command": "python",
      "args": ["-m", "sec_edgar_mcp.server"],
      "env": {
        "SEC_EDGAR_USER_AGENT": "Your Name (your.email@example.com)"
      }
    }
  }
}

If you used uvx:
{
  "mcpServers": {
    "sec-edgar-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/stefanoamorelli/sec-edgar-mcp.git",
        "sec-edgar-mcp"
      ],
      "env": {
        "SEC_EDGAR_USER_AGENT": "Your Name (your.email@example.com)"
      }
    }
  }
}

⚠️ IMPORTANT: Replace "Your Name (your.email@example.com)" with YOUR actual name and email.
The SEC requires this information and may block requests without proper identification.

Step 4: Test Your Connection

Restart Claude Desktop (or your AI client), then try this prompt:

"Can you use the SEC EDGAR tools to:
1. Look up Apple's CIK number
2. Get their most recent 10-K filing
3. Extract their total revenue for the latest year

Please cite the filing accession number."

Expected Result:
✓ AI discovers SEC EDGAR tools automatically
✓ Retrieves accurate filing data with accession numbers
✓ Provides financial figures from official SEC filings

Tools You'll See:
- lookup_company_cik (find company identifiers)
- get_recent_filings (retrieve 10-K, 10-Q, 8-K filings)
- get_financial_statements (extract financial data)
- get_company_facts (access XBRL tagged data)
- get_insider_transactions (track Form 3/4/5 filings)

Track SEC filings for data extraction - AI Infrastructure Ecosystem

AI CHIP MAKERS:
Company | CIK | Fiscal Year End | Filing Type | File Date | Accession Number
----------------------------------------------------------------------
NVIDIA | 0001045810 | January | 10-K | [Date] | [Number]
AMD | 0000002488 | December | 10-K | [Date] | [Number]
Intel | 0000050863 | December | 10-K | [Date] | [Number]
Taiwan Semi (TSMC) | 0001046179 | December | 20-F | [Date] | [Number]

CLOUD HYPERSCALERS:
Company | CIK | Fiscal Year End | Filing Type | File Date | Accession Number
----------------------------------------------------------------------
Microsoft | 0000789019 | June | 10-K | [Date] | [Number]
Amazon (AWS) | 0001018724 | December | 10-K | [Date] | [Number]
Alphabet (Google) | 0001652044 | December | 10-K | [Date] | [Number]
Meta (Facebook) | 0001326801 | December | 10-K | [Date] | [Number]

Note: This portfolio captures the entire AI infrastructure value chain:
- Chip Makers: Design and manufacture AI accelerators (GPUs, TPUs, custom silicon)
- Hyperscalers: Operate massive data centers, consume chips, provide AI services

Standard ratio definitions to implement in Excel

LIQUIDITY RATIOS:
• Current Ratio = Current Assets / Current Liabilities
• Quick Ratio = (Current Assets - Inventory) / Current Liabilities
• Cash Ratio = Cash & Equivalents / Current Liabilities

PROFITABILITY RATIOS:
• Gross Margin = Gross Profit / Revenue
• Operating Margin = Operating Income / Revenue
• Net Margin = Net Income / Revenue
• Return on Assets (ROA) = Net Income / Average Total Assets
• Return on Equity (ROE) = Net Income / Average Shareholders' Equity

EFFICIENCY RATIOS:
• Asset Turnover = Revenue / Average Total Assets
• Inventory Turnover = COGS / Average Inventory
• Days Sales Outstanding (DSO) = (Accounts Receivable / Revenue) × 365

LEVERAGE RATIOS:
• Debt-to-Equity = Total Debt / Total Equity
• Debt-to-Assets = Total Debt / Total Assets
• Interest Coverage = EBIT / Interest Expense

Structure for documenting and prioritizing risks

ACCOUNTING RISKS:
Risk Category | Description | Impact (H/M/L) | Likelihood (H/M/L) | Mitigation
---------------------------------------------------------------------------------
Revenue Recognition | Complex timing issues | High | Medium | [Analysis]
Estimates & Judgments | Warranty, bad debt | Medium | Medium | [Analysis]
Valuation | Goodwill, intangibles | High | Low | [Analysis]
Going Concern | Liquidity pressures | High | Low | [Analysis]

BUSINESS/OPERATIONAL RISKS:
Risk Category | Description | Impact (H/M/L) | Likelihood (H/M/L) | Mitigation
---------------------------------------------------------------------------------
Geopolitical | Taiwan tensions, trade | High | Medium | [Analysis]
Supply Chain | Fab capacity, materials | High | Medium | [Analysis]
Technology | AI transition, Moore's Law | High | High | [Analysis]
Competition | AMD, Intel market share | Medium | High | [Analysis]
Regulatory | Export controls, antitrust | Medium | Medium | [Analysis]

Copy these prompts to your AI tool to generate forecasts

PROMPT 1: Historical Analysis
"I have 5 years of financial data for NVIDIA (2019-2023). 
Revenues: [paste your data]
Gross Margins: [paste your data]

Please:
1. Calculate the revenue CAGR
2. Identify revenue growth trends and any patterns
3. Analyze margin trends and key drivers
4. Summarize any significant changes year-over-year"

---

PROMPT 2: Scenario Generation
"Based on this historical data and semiconductor industry trends (AI boom, geopolitical risks, cyclical nature), please create three 3-year forecast scenarios for NVIDIA:

1. BASE SCENARIO (most likely)
2. OPTIMISTIC SCENARIO (AI demand accelerates)
3. PESSIMISTIC SCENARIO (cyclical downturn)

For each scenario, provide:
- Revenue growth assumptions with justification
- Gross margin assumptions
- Key drivers and risks
- Year-by-year projections (2024-2026)"

---

PROMPT 3: Sensitivity Analysis
"Using the Base scenario, perform sensitivity analysis:

1. Test revenue growth at -10%, -5%, 0%, +5%, +10% from base
2. Test gross margin at -3%, -2%, -1%, +1%, +2%, +3% from base
3. Create a sensitivity table showing impact on operating income
4. Identify which variables have the highest impact"

---

PROMPT 4: Cross-Check & Validation
"Review these forecast assumptions against:
1. Historical performance patterns
2. Industry analyst expectations
3. Management guidance from recent 10-K/earnings calls
4. Peer company trends

Highlight any assumptions that seem unrealistic or need adjustment."

Comprehensive report organization

# AI-Powered Semiconductor Industry Analysis: NVIDIA & Peer Comparative Study

## Executive Summary (≤500 words)
- Project scope and methodology overview
- Key findings: financial health, competitive positioning
- Critical risks and opportunities identified
- Forecast summary: base case with scenario ranges
- Primary recommendations

## 1. Introduction & Methodology
- 1.1 Project Objectives
- 1.2 Company Universe (7 companies, rationale)
- 1.3 Data Sources (SEC filings, accession numbers)
- 1.4 Analysis Approach (AI tools used, validation methods)
- 1.5 Limitations & Assumptions

## 2. Company Profiles & Historical Performance
- 2.1 NVIDIA (Primary Focus)
- 2.2 AMD, TSM, Intel (Brief profiles)
- 2.3 Micron, Broadcom, Marvell (Brief profiles)
- 5-year revenue, margin, profitability trends

## 3. Financial Ratio Analysis
- 3.1 Liquidity Analysis (Current, Quick, Cash ratios)
- 3.2 Profitability Analysis (Margins, ROA, ROE)
- 3.3 Efficiency Analysis (Turnover ratios, DSO)
- 3.4 Leverage Analysis (Debt ratios, coverage)
- 3.5 Cross-Company Comparison & Rankings

## 4. Industry Intelligence & Competitive Landscape
- 4.1 Semiconductor Industry Overview
- 4.2 AI Infrastructure Market Dynamics
- 4.3 Supply Chain Analysis
- 4.4 Competitive Positioning Matrix
- 4.5 Technology Trends & Disruptions

## 5. Risk Assessment
- 5.1 Accounting Risks (Revenue, Estimates, Valuation)
- 5.2 Business/Operational Risks (Geopolitical, Market, Tech)
- 5.3 Risk Prioritization Matrix
- 5.4 Mitigation Strategies

## 6. Forecasting & Scenario Analysis
- 6.1 Forecast Methodology & Assumptions
- 6.2 Base Scenario (3-year projections)
- 6.3 Optimistic Scenario
- 6.4 Pessimistic Scenario
- 6.5 Sensitivity Analysis
- 6.6 Scenario Comparison & Implications

## 7. Conclusions & Recommendations
- 7.1 Key Takeaways
- 7.2 Investment/Strategic Implications
- 7.3 Areas for Further Investigation
- 7.4 Lessons Learned (AI workflow effectiveness)

## Appendices
- A. Ratio Definitions & Formulas
- B. Detailed Assumption Register
- C. Data Lineage & Source Documentation
- D. Prompt Log (AI tools used)
- E. Validation Checklist Results
- F. References & Citations

Recommended slide organization for executive deck

SLIDE 1: Title Slide
- Project title, your name, date

SLIDE 2: Executive Summary
- Project scope, key findings (bullet points)

SLIDE 3: Methodology Overview
- Data sources, analysis approach, AI tools used

SLIDE 4: Company Universe
- 7 companies analyzed, selection rationale

SLIDE 5-6: Historical Performance
- 5-year revenue and margin trends (charts)
- Key performance highlights

SLIDE 7-8: Financial Ratio Analysis
- Ratio comparison tables/charts
- Peer ranking and competitive insights

SLIDE 9: Industry Context
- Semiconductor market dynamics
- AI infrastructure growth drivers

SLIDE 10-11: Risk Assessment
- Risk matrix visualization
- Top 5 critical risks with mitigation

SLIDE 12-14: Forecast Scenarios
- Base/Optimistic/Pessimistic scenario comparison
- Sensitivity analysis table
- Key assumptions documented

SLIDE 15: Conclusions & Recommendations
- Primary takeaways (3-5 bullets)
- Strategic implications

SLIDE 16: Q&A Preparation
- Thank you + contact information

APPENDIX SLIDES (as needed):
- Detailed ratio definitions
- Additional charts/tables
- Methodology details

Core governance artifacts to maintain throughout project

PROMPT LOG FORMAT:
Date | AI Tool | Purpose | Prompt Text | Output Summary | Validation Status | Notes
-----------------------------------------------------------------------------------------
[Date] | Claude Desktop | Extract NVIDIA revenue | [Full prompt] | [Key results] | Verified ✓ | Spot-checked against 10-K

VALIDATION CHECKLIST FORMAT:
Phase | Item Checked | Method | Date | Result | Issues Found | Resolution
-----------------------------------------------------------------------------
Phase 1 | NVIDIA 2023 revenue | Manual 10-K lookup | [Date] | Match ✓ | None | N/A
Phase 2 | Current Ratio formula | Excel audit | [Date] | Correct ✓ | None | N/A

ASSUMPTIONS REGISTER FORMAT:
Assumption | Value/Range | Justification | Source/Citation | Risk Level | Sensitivity
------------------------------------------------------------------------------------------
Base revenue CAGR | 25% | Historical 3Y average + industry outlook | Company filings + research | Medium | High

DATA LINEAGE REGISTER FORMAT:
Metric | Company | Year | Value | Source Filing | Page/Section | Accession # | Verified ✓/✗
----------------------------------------------------------------------------------------------
Revenue | NVIDIA | 2023 | $26.97B | 10-K | Pg 51, IS | 0001045810-24-000029 | ✓

CHANGE LOG FORMAT:
Date | Phase | Change Type | Description | Justification | Approver | Impact
--------------------------------------------------------------------------------
[Date] | Phase 3 | Scope adjustment | Added ESG risk factor | Found material disclosure | [Name] | Minor