Skip to main content
command CX Codex official · Anthropic

dcf — Claude Code command: DCF valuation and setup

This Claude Code command builds a company DCF valuation and checks it against public peers. It uses comparable-company multiples to shape valuation assumptions.

Apache-2.0 ★ 41,441 updated 2026-09-21

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add anthropics/financial-services
/plugin install financial-analysis@claude-for-financial-services
Open on GitHub

What dcf does

The command prepares a discounted cash flow valuation and compares its assumptions with public peers. It first calls the comps-analysis skill to find 4–6 comparable companies, collect financial metrics and summarize their trading multiples.

The dcf-model skill then builds Bear, Base and Bull cases, calculates WACC with CAPM, and derives equity value and an implied share price. Deliverables include peer analysis and a DCF model in .xlsx, sensitivity tables, and a summary of valuation drivers and risks.

The workflow requires the comps-analysis and dcf-model skills, plus historical company financials and market data.

What it helps with

  • compare a DCF valuation with public-company trading multiples
  • estimate implied share value across Bear, Base and Bull cases
  • cross-check implied EV/EBITDA and P/E against peers
  • export peer analysis and the DCF model as .xlsx files

Example prompts for the agent

  • /financial-analysis:dcf for Microsoft
  • /financial-analysis:dcf value AAPL using a DCF model

How to install dcf

1Plugin marketplace

/plugin marketplace add anthropics/financial-services
/plugin install financial-analysis@claude-for-financial-services

2Manually: put the files in place

.claude/commands/dcf.md

Call it with /dcf. For all projects use ~/.claude/commands/

Questions about dcf

How do I install dcf in Claude Code?

Run /plugin marketplace add anthropics/financial-services, then /plugin install financial-analysis@claude-for-financial-services. The command file can also be installed at .claude/commands/dcf.md.

What does dcf do?

It builds a DCF valuation, uses public-company multiples to inform assumptions, and checks the result against peer metrics. The output includes .xlsx files and multiple valuation cases.

Which skills does dcf require?

It needs comps-analysis for peer analysis and dcf-model to construct the DCF valuation.

Source text

dcf.md · Apache-2.0
---
description: Build a DCF valuation model with comps-informed terminal multiples
argument-hint: "[company name or ticker]"
---

# DCF Valuation Command

Build an institutional-quality DCF model that uses comparable company analysis to inform valuation ranges.

## Workflow

### Step 1: Gather Company Information

If a company name or ticker is provided, use it. Otherwise ask:
- "What company would you like to value?"

### Step 2: Run Comparable Company Analysis

**First, load the comps-analysis skill** to build trading comps:

Use `skill: "comps-analysis"` to:
1. Identify 4-6 comparable public companies
2. Pull operating metrics (Revenue, EBITDA, margins, growth)
3. Pull valuation multiples (EV/Revenue, EV/EBITDA, P/E)
4. Calculate statistical summary (median, 25th/75th percentiles)

**Key outputs to capture from comps:**
- Median EV/EBITDA multiple → informs terminal value exit multiple
- Median EV/Revenue multiple → sanity check on DCF output
- Peer growth rates → benchmark for revenue projections
- Peer margins → benchmark for margin assumptions

### Step 3: Build DCF Model

**Load the dcf-model skill** to construct the valuation:

Use `skill: "dcf-model"` to:
1. Gather historical financials and market data
2. Build revenue projections (Bear/Base/Bull cases)
3. Model operating expenses and FCF
4. Calculate WACC using CAPM
5. Discount cash flows and calculate terminal value
6. Bridge to equity value and implied share price

**Use comps to inform DCF assumptions:**

| Comps Output | DCF Input |
|--------------|-----------|
| Peer median EV/EBITDA | Terminal exit multiple range |
| Peer 25th-75th EV/EBITDA | Sensitivity analysis range |
| Peer median growth rate | Benchmark for revenue assumptions |
| Peer median EBITDA margin | Target margin in terminal year |
| Peer median P/E | Cross-check implied P/E from DCF |

### Step 4: Cross-Check Valuation

After DCF is complete, validate:
1. **Implied EV/EBITDA** from DCF vs peer median
   - If DCF implies 25x but peers trade at 12x, investigate why
2. **Implied P/E** from DCF vs peer median
3. **Terminal value as % of EV** (should be 50-70%)
4. **Implied growth** embedded in valuation vs peer growth rates

### Step 5: Deliver Output

Provide:
1. **Comps analysis spreadsheet** (.xlsx) with peer trading multiples
2. **DCF model** (.xlsx) with:
   - Bear/Base/Bull scenarios
   - Sensitivity tables (WACC vs Terminal Growth, etc.)
   - Valuation summary with implied upside/downside
3. **Summary** explaining:
   - Key valuation drivers
   - How comps informed the analysis
   - Risks and sensitivities to watch

## Example Output Summary

```
VALUATION SUMMARY: [Company] ([Ticker])

Comparable Companies Analysis:
- Peer Group: [List of 4-6 comps]
- Median EV/EBITDA: 12.5x (range: 10.2x - 15.8x)
- Median EV/Revenue: 3.2x (range: 2.1x - 4.5x)

DCF Valuation (Base Case):
- Implied Share Price: $XX.XX
- Current Price: $YY.YY
- Implied Upside: +XX%

Valuation Cross-Check:
- DCF Implied EV/EBITDA: 13.2x (vs peer median 12.5x)
- DCF Implied P/E: 22.4x (vs peer median 20.1x)
- Terminal Value: 62% of EV (within normal range)

Key Assumptions:
- Revenue CAGR: X% (vs peer median X%)
- Terminal EBITDA Margin: X% (vs peer median X%)
- WACC: X.X%
- Terminal Growth: X.X%
```

© Anthropic FSI · Apache-2.0 · anthropics/financial-services

Similar slash commands

All slash commands for Finance and business →

From the same set116

The whole set →
ai-readiness The ai-readiness Claude Code command evaluates AI opportunities across portfolio companies. It ranks quick wins by EBITDA impact and spots initiatives that could apply to multiple companies. CX Codex +1 ★ 41,441repo stars one-pager The one-pager Claude Code command creates a one-page company profile for pitch books and deal materials. Give it a company name or ticker to assemble a PowerPoint slide with business and financial details. CX Codex +1 ★ 41,441repo stars ppt-template The ppt-template Claude Code command turns a PowerPoint template into a reusable skill. It inspects a .pptx or .potx file, builds the skill package, and checks it with a sample deck. CX Codex +1 ★ 41,441repo stars debug-model debug-model — a Claude Code command that checks financial models for formula and logic errors. Run it to audit an .xlsx model for balance issues, hardcoded overrides, and broken references. CX Codex +1 ★ 41,441repo stars comps The comps Claude Code command builds a comparable-company analysis using operating metrics and valuation multiples. Give it a company name or ticker to generate an Excel workbook and a summary. CX Codex +1 ★ 41,441repo stars earnings earnings — a command for Claude Code, Codex, GitHub Copilot and Gemini CLI that compiles quarterly financial results and generates an equity research report in DOCX. CX Codex +1 ★ 41,441repo stars review-fi-portfolio The review-fi-portfolio Claude Code command reviews fixed-income portfolios using prices, cashflows, and rate scenarios. Supply bond identifiers to start the analysis. CX Codex +1 ★ 41,441repo stars research-equity research-equity is a Claude Code command that compiles equity research summaries using consensus estimates, company fundamentals, and price data. Runs when analyzing a stock by ticker. CX Codex +1 ★ 41,441repo stars

Project sponsors

Companies that keep this analytics open