excel-analyst-pro
by jeremylongshorev1.0.0
Professional financial modeling toolkit for Claude Code with auto-invoked Skills and Excel MCP integration. Build DCF models, LBO analysis, variance reports, and pivot tables using natural language.
Documentation
# Excel Analyst Pro
**Professional financial modeling toolkit for Claude Code with auto-invoked Skills and Excel MCP integration.**
Build DCF models, LBO analysis, variance reports, and pivot tables using natural language. No formulas to remember, no manual Excel work—just describe what you need.
---
## 🚀 Features
### Auto-Invoked Skills
- **DCF Modeler**: Build discounted cash flow valuation models with projections, WACC, and sensitivity analysis
- **LBO Modeler**: Create leveraged buyout models with debt schedules, cash flow waterfalls, and IRR calculations
- **Variance Analyzer**: Automate budget vs actual analysis with flagging, commentary, and executive summaries
- **Pivot Wizard**: Generate pivot tables and charts from raw data using natural language
### Key Benefits
- ✅ **Natural Language**: Just describe what you want—no Excel formulas to remember
- ✅ **Auto-Invoked**: Claude automatically detects when to use each Skill
- ✅ **Investment Banking Grade**: Follows IB best practices and standards
- ✅ **Local Processing**: No cloud upload—all Excel work happens locally
- ✅ **Token Efficient**: Skills load only when relevant (50% cheaper than MCP alone)
- ✅ **Team Standardization**: Everyone uses the same best practices
---
## 📦 Installation
### Prerequisites
- Claude Code 1.0+
- Node.js 18+
### Quick Install
```bash
/plugin install excel-analyst-pro@claude-code-plugins
```
This will automatically:
1. Install the plugin
2. Configure @negokaz/excel-mcp-server
3. Load all 4 Skills into Claude Code
4. Add slash commands
---
## 🎯 Usage
### DCF Valuation Model
Just ask in natural language:
```
You: "Create a DCF model for Tesla"
Claude: I'll build a DCF valuation model for Tesla. Let me gather some inputs.
What base year revenue should we use for Tesla?
(For reference, Tesla's 2023 revenue was approximately $96.8B)
You: $96.8 billion
Claude: Great! What revenue growth rates should we assume for Years 1-5?
You: 25%, 20%, 15%, 12%, 10%
Claude: [Builds complete DCF model with 4 sheets]
✅ DCF Model Complete!
📊 Valuation Results:
- Enterprise Value: $847.3 billion
- Implied equity value per share: $243
📁 Model saved to: Tesla_DCF_Model.xlsx
```
**That's it!** No Excel formulas, no manual work.
---
### LBO Analysis
```
You: "Build an LBO model for a $50M EBITDA software company"
Claude: I'll create an LBO model. What entry multiple should we assume?
(Software companies typically trade at 10-15x EBITDA)
You: 12x
Claude: [Builds complete LBO model with 6 sheets]
✅ LBO Model Complete!
📊 Base Case Returns (5-year hold, 12x exit):
- Exit Equity Value: $1,124M
- Money-on-Money: 4.2x
- IRR: 34.2%
📁 Model saved to: Software_LBO_Model.xlsx
```
---
### Variance Analysis
```
You: "Analyze Q1 budget vs actual"
Claude: [Analyzes data and generates report]
✅ Variance Analysis Complete!
🔴 CRITICAL VARIANCES:
- EBITDA: $270K vs $450K (-40.0%) - $180K under
- Operating Expenses: $840K vs $750K (+12.0%) - $90K over
📁 Report saved to: Q1_2025_Variance_Analysis.xlsx
```
---
### Pivot Tables
```
You: "Show sales by region and product category"
Claude: [Creates pivot table and visualization]
✅ Pivot Table Created!
Top Insights:
1. West region leads with $1.45M (30.3% of total)
2. Electronics is #1 category at $1.70M (35.6%)
3. Northeast + Electronics = $400K (highest cell)
Visualization: Added column chart comparing regions
```
---
## FREE Financial Data Sources: No Bloomberg Required
**Get company financials, market data, and economic indicators for your models** - all free, no $24K/year Bloomberg subscription.
### Quick Comparison
| Data Type | Paid Source | FREE Source |
|-----------|-------------|-------------|
| **Company Financials** | Bloomberg ($24K/year) | SEC EDGAR: **$0** |
| **Stock Prices** | Capital IQ ($12K/year) | Yahoo Finance: **$0** |
| **Market Data** | FactSet ($12K/year) | Alpha Vantage: **$0** |
| **Macro Indicators** | Refinitiv ($12K/year) | FRED: **$0** |
| **Company News** | S&P CapitalIQ ($12K/year) | Google News: **$0** |
**Annual Savings: $25K-74K** for professional-grade model inputs.
### Why Free Data Works for Financial Modeling
**For DCF Models:**
- Revenue/EBITDA: SEC 10-K/10-Q filings (FREE)
- Stock prices: Yahoo Finance (FREE)
- Risk-free rate: FRED (Federal Reserve, FREE)
- Beta: Calculated from Yahoo Finance data (FREE)
**For LBO Models:**
- Entry valuation: SEC filings + Yahoo Finance (FREE)
- Debt terms: Company 10-K disclosure (FREE)
- Comparable multiples: Public comps from Yahoo Finance (FREE)
- Exit assumptions: Historical trading multiples (FREE)
**For Variance Analysis:**
- Budget data: Your internal files (already have)
- Actual results: Your accounting system (already have)
- Industry benchmarks: BEA.gov, Census.gov (FREE)
**15-minute delayed data is perfectly fine for financial modeling** (not day trading).
### Free Data Source Catalog
#### 1. SEC EDGAR (Best for Fundamentals)
**What:** Official company filings (10-K, 10-Q, 8-K)
**Use For:**
- Revenue, EBITDA, net income
- Balance sheet data
- Cash flow statements
- Management discussion & analysis (MD&A)
- Risk factors
**Access:**
- Website: [sec.gov/edgar](https://www.sec.gov/edgar)
- API: FREE, unlimited access
- Python: `pip install sec-api` (free tier)
**Cost:** $0 (US government public data)
**Example:**
```python
# Get Tesla's latest 10-K
import requests
cik = "0001318605" # Tesla's CIK
url = f"https://data.sec.gov/submissions/CIK{cik}.json"
response = requests.get(url, headers={"User-Agent": "YourName [email protected]"})
filings = response.json()
```
#### 2. Yahoo Finance (Best for Stock Data)
**What:** Real-time stock prices, historical data, key stats
**Use For:**
- Current stock price
- Historical prices (for beta calculation)
- Market cap
- P/E, EV/EBITDA ratios
- 52-week high/low
**Access:**
- Website: [finance.yahoo.com](https://finance.yahoo.com)
- Python: `pip install yfinance` (FREE)
- Excel: Power Query (built-in, FREE)
**Cost:** $0
**Example:**
```python
import yfinance as yf
# Get Tesla data for DCF model
tesla = yf.Ticker("TSLA")
revenue = tesla.financials.loc["Total Revenue"]
stock_price = tesla.history(period="1d")["Close"].iloc[0]
market_cap = tesla.info["marketCap"]
```
#### 3. FRED (Federal Reserve Economic Data)
**What:** 817,000+ economic time series
**Use For:**
- Risk-free rate (10-year Treasury)
- GDP growth rates
- Inflation (CPI)
- Unemployment rates
- Market risk premium data
**Access:**
- Website: [fred.stlouisfed.org](https://fred.stlouisfed.org)
- API: FREE (no rate limits)
- Excel: FRED Excel Add-in (FREE)
**Cost:** $0
**Example:**
```python
from fredapi import Fred
fred = Fred(api_key="YOUR_FREE_KEY") # Free key from FRED website
# Get 10-year Treasury rate for WACC calculation
risk_free_rate = fred.get_series_latest_release("DGS10")
print(f"Current risk-free rate: {risk_free_rate.iloc[-1]}%")
```
#### 4. Alpha Vantage (Best for Technicals)
**What:** Stock fundamentals, technical indicators, forex
**Use For:**
- Financial statements (income statement, balance sheet, cash flow)
- Key ratios
- Earnings calendar
- Technical indicators (SMA, RSI)
**Access:**
- Website: [alphavantage.co](https://www.alphavantage.co)
- API: FREE tier (500 calls/day)
**Cost:** $0 (free tier sufficient for modeling)
**Example:**
```python
import requests
api_key = "YOUR_FREE_KEY" # Free from alphavantage.co
symbol = "AAPL"
url = f"https://www.alphavantage.co/query?function=INCOME_STATEMENT&symbol={symbol}&apikey={api_key}"
response = requests.get(url)
financials = response.json()
```
#### 5. OpenBB Platform (Best All-in-One)
**What:** Unified interface to 100+ free data providers
**Use For:**
- Stocks, crypto, forex, commodities
- Fundamentals, technicals, macro
- Portfolio analytics
**Install:** `pip install openbb[yfinance]`
**Cost:** $0 (uses free providers)
**See:** [openbb-terminal plugin](../../finance/openbb-terminal/) for full guide
### Cost Comparison: Building a DCF Model
#### Paid Data Approach
**Annual Subscriptions:**
- Bloomberg Terminal: $24,000/year
- Capital IQ: $12,000/year
- FactSet: $12,000/year
- **Total: $48,000/year**
**Advantages:**
- Real-time data
- Instant analyst estimates
- Proprietary research
#### Free Data Approach
**Annual Subscriptions:**
- SEC EDGAR: $0
- Yahoo Finance: $0
- FRED: $0
- Alpha Vantage: $0
- **Total: $0/year**
**Advantages:**
- Same official company data (SEC filings)
- 15-min delayed (fine for modeling)
- No credit card required
**Savings: $48,000/year** with identical model quality.
### Real Use Case Examples
#### DCF Model for Apple
**Paid Approach (Bloomberg):**
1. Open Bloomberg Terminal ($24K/year)
2. Type `AAPL <EQUITY> FA` for financials
3. Export to Excel
4. Build DCF model
**Free Approach (This Plugin):**
```python
import yfinance as yf
# Get Apple data
aapl = yf.Ticker("AAPL")
revenue = aapl.financials.loc["Total Revenue"]
operating_income = aapl.financials.loc["Operating Income"]
market_cap = aapl.info["marketCap"]
# Get risk-free rate from FRED
from fredapi import Fred
fred = Fred(api_key="YOUR_FREE_KEY")
risk_free_rate = fred.get_series_latest_release("DGS10").iloc[-1]
# Now use Excel Analyst Pro to build DCF
# Just say: "Create a DCF model for Apple"
```
**Cost:** $0 (vs $24K/year)
**Data Quality:** Identical (both use SEC filings + public market data)
#### LBO Model for Private Company
**Data Needed:**
- Entry valuation: Ask seller or use industry multiples
- Debt terms: Term sheets from lenders
- EBITDA projections: Internal management projections
- Exit assumptions: Public comparable multiples (Yahoo Finance)
**Cost with Free Data:** $0
**No paid subscriptions required for private company LBO models.**
#### Variance Analysis
**Data Needed:**
- Budget: Your internal Excel file
- Actuals: Your accounting system export
- KPIs: Your tracking dashboards
**Cost:** $0 (all internal data)
### Integration with This Plugin
**Step 1:** Get free data from sources above
**Step 2:** Use Excel Analyst Pro to build models
```
You: "Create a DCF model for Tesla"
Claude: What base year revenue should we use for Tesla?
You: "$96.8 billion" (from SEC 10-K or Yahoo Finance, both FREE)
Claude: [Builds complete DCF model]
```
**Step 3:** Save $48K/year by avoiding Bloomberg
### When Free Data Is NOT Enough
**Use paid data if:**
- You're an investment bank pitching M&A ($24K/year justified)
- You need real-time intraday data for trading
- Client requires Bloomberg screenshots for compliance
- You manage $1B+ AUM and need institutional tools
**For everyone else (99% of users):** Free data is sufficient for professional financial models.
### Hybrid Approach
**Best of both worlds:** Use free data 95% of the time, Bloomberg for final client deliverables.
**Cost Reduction:** $48K/year → $2.4K/year (95% savings)
### Resources
- **SEC EDGAR:** [sec.gov/edgar](https://www.sec.gov/edgar) (FREE)
- **Yahoo Finance:** [finance.yahoo.com](https://finance.yahoo.com) (FREE)
- **FRED:** [fred.stlouisfed.org](https://fred.stlouisfed.org) (FREE API key)
- **Alpha Vantage:** [alphavantage.co](https://www.alphavantage.co) (FREE API key)
- **OpenBB:** Install [openbb-terminal plugin](../../finance/openbb-terminal/) for unified access
**Bottom Line:** This plugin is free. Your model inputs can be free too. Save $25K-74K/year.
---
## 📚 Skills Documentation
Each Skill has detailed documentation in its `SKILL.md` file:
### 1. DCF Modeler
**File:** `skills/excel-dcf-modeler/SKILL.md`
**Triggers:**
- "Create a DCF model"
- "Build a valuation model"
- "Calculate enterprise value"
- "Value [company]"
**Outputs:**
- 4-sheet Excel model (Assumptions, FCF Projections, Valuation, Sensitivity)
- Enterprise value calculation
- Sensitivity analysis (WACC vs terminal growth)
---
### 2. LBO Modeler
**File:** `skills/excel-lbo-modeler/SKILL.md`
**Triggers:**
- "Create an LBO model"
- "Build a leveraged buyout model"
- "Private equity analysis"
- "Calculate IRR for acquisition"
**Outputs:**
- 6-sheet Excel model (Transaction, Sources & Uses, Operating, Debt Schedule, Returns, Covenants)
- IRR and money-on-money calculations
- Multiple sensitivity tables
---
### 3. Variance Analyzer
**File:** `skills/excel-variance-analyzer/SKILL.md`
**Triggers:**
- "Analyze budget variance"
- "Compare actual vs forecast"
- "Create variance report"
- "Why are we over/under budget?"
**Outputs:**
- 3-sheet Excel report (Variance Summary, Executive Summary, Trend Analysis)
- Automated flagging (🔴 critical, ⚠️ warning, ✅ on track)
- Commentary and recommendations
---
### 4. Pivot Wizard
**File:** `skills/excel-pivot-wizard/SKILL.md`
**Triggers:**
- "Create a pivot table"
- "Analyze [data] by [dimension]"
- "Summarize sales by region"
- "Show revenue breakdown"
**Outputs:**
- Pivot tables with professional formatting
- Charts and visualizations
- Slicers and filters
- Calculated fields
---
## 🛠️ Technical Details
### Plugin Architecture
```
excel-analyst-pro/
├── plugin.json # Plugin configuration
├── README.md # This file
├── skills/ # Auto-invoked Skills
│ ├── excel-dcf-modeler/
│ │ ├── SKILL.md # DCF modeling instructions
│ │ └── resources/
│ │ └── dcf-template.xlsx
│ ├── excel-lbo-modeler/
│ ├── excel-variance-analyzer/
│ └── excel-pivot-wizard/
├── mcp/ # MCP server config
│ └── excel-config.json
├── slash-commands/ # Manual triggers (optional)
│ ├── build-dcf.md
│ ├── build-lbo.md
│ └── analyze-variance.md
└── examples/ # Example files
```
### MCP Server Integration
This plugin uses `@negokaz/excel-mcp-server` for Excel operations:
**Configuration:**
```json
{
"command": "npx",
"args": ["--yes", "@negokaz/excel-mcp-server"],
"env": {
"EXCEL_MCP_PAGING_CELLS_LIMIT": "4000"
}
}
```
**Capabilities:**
- Read and write Excel (.xlsx) files
- Create sheets and workbooks
- Write formulas
- Format cells
- No Microsoft Excel installation required
---
## 💡 How It Works
### Skills + MCP Server = Complete Solution
**MCP Server (Low-Level):**
- Provides Excel file operations
- Read cells, write formulas, create sheets
- Like having a toolbox 🧰
**Skills (High-Level):**
- Provide domain expertise
- Financial modeling patterns, best practices
- Like having an expert consultant 👔
**Together:**
- Professional results with zero effort 🚀
### Example Flow
```
User: "Create a DCF model"
↓
Claude detects "DCF model" in request
↓
Auto-loads excel-dcf-modeler Skill
↓
Skill provides instructions:
- Ask for company and assumptions
- Build 4-sheet model structure
- Calculate enterprise value
- Create sensitivity tables
↓
Skill calls MCP server to:
- create_workbook()
- add_sheet()
- write_formula()
- format_cells()
↓
Output: Professional DCF model
```
**No manual invocation needed!** Just natural language.
---
## 📈 Use Cases
### Investment Banking Analysts
- **Before**: 2-3 hours to build DCF from scratch
- **After**: 5 minutes with professional results
- **Time Saved**: 2-3 hours per model
### Private Equity Associates
- **Before**: Full day to build LBO model
- **After**: 10 minutes with complete analysis
- **Time Saved**: 7-8 hours per deal
### Finance Managers
- **Before**: 2 hours for monthly variance analysis
- **After**: 5 minutes with automated commentary
- **Time Saved**: 1 hour 55 minutes per month
### Data Analysts
- **Before**: 55 minutes to clean data, create pivots, format charts
- **After**: 3 minutes end-to-end
- **Time Saved**: 52 minutes per report
---
## 🎓 Best Practices
### 1. Start with Defaults
Don't have all assumptions? No problem:
```
You: "Build a DCF but I don't have all the numbers"
Claude: No problem! I can build with industry-average assumptions.
What company are you valuing?
You: Amazon
Claude: I'll use standard assumptions for large-cap tech:
- Revenue growth: 15%, 12%, 10%, 8%, 6%
- EBITDA margin: 25%
- WACC: 9%
You can adjust the "Assumptions" sheet later.
```
### 2. Ask for Explanations
```
You: "Explain the DCF model you built"
Claude: Sure! Let me walk through the DCF model:
📊 What is a DCF?
A DCF values a company based on its future cash flows...
[Detailed explanation provided]
```
### 3. Iterate and Refine
```
You: "The WACC seems high—can we try 8% instead of 10%?"
Claude: Absolutely! Updating WACC from 10% to 8%...
New Enterprise Value: $924B (was $847B)
Change: +$77B (+9.1%)
Lower discount rate increases value as expected.
```
### 4. Request Variations
```
You: "Add a dividend recap in Year 3 to the LBO model"
Claude: I'll add a dividend recapitalization to the model.
In Year 3, after debt paydown:
- Refi to 5.0x EBITDA (from 3.8x)
- Distribute ~$150M to equity sponsors
This boosts IRR by ~300bps with partial liquidity.
```
---
## ⚠️ Limitations
### What This Plugin Does Well
- Initial valuation analysis ✅
- Pitch decks and presentations ✅
- Academic exercises ✅
- Quick "back of envelope" valuations ✅
- Team standardization ✅
### What Requires Manual Work
- Official fairness opinions ❌
- Detailed IC presentations (need scenario planning) ❌
- Quarterly debt schedules (annual only) ❌
- Complex waterfall structures ❌
- Third-party data validation ❌
**Recommendation:** Use this plugin for initial analysis (80% of use cases), then refine manually for high-stakes decisions.
---
## 🔒 Security & Privacy
### Local Processing
- ✅ All Excel work happens locally
- ✅ No cloud upload required
- ✅ Works offline
- ✅ Full control over data
**vs Claude for Excel (cloud-based):**
- ❌ Uploads data to Anthropic cloud
- ❌ Requires Max/Enterprise/Teams subscription
- ❌ Subject to data retention policies
**Recommendation:** For sensitive financial data or regulated industries, use this plugin (local processing) instead of cloud-based solutions.
---
## 🆚 Comparison
| Feature | Excel Analyst Pro | Claude for Excel | Microsoft Copilot |
|---------|------------------|------------------|-------------------|
| **Price** | Free (open-source) | Max/Enterprise subscription | $30/user/month |
| **Processing** | Local | Cloud | Cloud |
| **Skills Included** | 4 (DCF, LBO, Variance, Pivot) | Limited beta | General assistance |
| **Customizable** | ✅ Fully | ❌ No | ❌ No |
| **Team Sharing** | ✅ Copy/paste Skills | ❌ Cloud only | ❌ Cloud only |
| **Financial Models** | ✅ IB-grade templates | ✅ (beta) | ❌ Basic |
| **Token Efficient** | ✅ Skills on-demand | ❌ Always loaded | N/A |
---
## 🐛 Troubleshooting
### Plugin Not Loading
```bash
# Check Claude Code version
claude --version
# Reinstall plugin
/plugin uninstall excel-analyst-pro
/plugin install excel-analyst-pro@claude-code-plugins
```
### MCP Server Not Working
```bash
# Verify Node.js version
node --version # Should be 18+
# Manually test MCP server
npx --yes @negokaz/excel-mcp-server
```
### Skill Not Triggering
**Problem:** You said "create DCF" but the Skill didn't load.
**Solution:** Be more explicit:
```
❌ "create DCF"
✅ "Create a DCF model for Apple"
```
Skills trigger on description matching—provide enough context.
---
## 🤝 Contributing
Want to add more Skills or improve existing ones?
1. Fork the repository
2. Create a new Skill in `skills/`
3. Follow the SKILL.md format (see existing Skills)
4. Test with real Excel workflows
5. Submit a pull request
**Ideas for new Skills:**
- Comparable company analysis (comps)
- M&A accretion/dilution model
- Three-statement financial model
- Portfolio performance tracker
- Budget template generator
---
## 📄 License
MIT License - see LICENSE file for details.
---
## 🙏 Acknowledgments
- **Anthropic** for Claude Code and Skills system
- **@negokaz** for excel-mcp-server
- **Investment banking community** for modeling best practices
---
## 📞 Support
- **Issues**: https://github.com/jeremylongshore/claude-code-plugins/issues
- **Discussions**: https://github.com/jeremylongshore/claude-code-plugins/discussions
- **Website**: https://claudecodeplugins.io
---
## 🚀 What's Next
### Roadmap
- [ ] Comparable company analysis Skill
- [ ] M&A accretion/dilution Skill
- [ ] Three-statement model builder
- [ ] Chart generation Skill
- [ ] VBA macro assistant
- [ ] Excel → Database migration tools
### Version History
- **v1.0.0** (2025-10-27): Initial release with 4 core Skills
---
**Made with ❤️ for the Claude Code community**
**Star this repo** if you find it useful! ⭐