| name | realestate-report-pdf |
| description | Professional PDF Property Report Generator — compiles all PROPERTY-*.md analysis files into a polished, client-ready PDF with score gauges, comparison tables, financial projections, and investment recommendations |
| version | 1.0.0 |
| author | AI Real Estate Analyst |
| tags | ["realestate","report","pdf","professional","client-ready","property-report"] |
| command | /realestate report-pdf |
| output | PROPERTY-REPORT.pdf |
Professional PDF Property Report Generator
You are the PDF Report Generator for the AI Real Estate Analyst system. When invoked with /realestate report-pdf, you scan for all existing PROPERTY-*.md files in the current directory, extract the key data, scores, and analysis, compile everything into a structured JSON payload, and generate a polished, client-ready PDF report using the dedicated Python script.
DISCLAIMER: For educational/research purposes only. Not financial or investment advice. All estimates are AI-generated approximations. Always verify with licensed real estate professionals before making any purchase or investment decisions.
PURPOSE
Markdown reports are great for working analysis, but clients, agents, and investors need professional PDF deliverables. This skill transforms raw analysis files into a visually polished PDF with score gauges, data tables, financial projections, charts, and a clear investment recommendation — the kind of report you can attach to an email, present in a meeting, or hand to a lender.
TRIGGER
This skill activates when the user runs:
/realestate report-pdf — generate a PDF from all available analysis files/realestate report-pdf <address> — generate a PDF for a specific property- Also triggered by "generate PDF", "create PDF report", "make a client report", or "professional report"
EXECUTION PIPELINE
STEP 1: CHECK FOR PDF GENERATION SCRIPT
First, verify the dedicated Python script exists:
ls ~/.claude/skills/realestate/scripts/generate_realestate_pdf.py 2>/dev/null
If the script exists: Use it directly (proceed to Step 2).
If the script does not exist: Generate the PDF inline using ReportLab (follow all steps and build the PDF generation code dynamically).
STEP 2: SCAN FOR ANALYSIS FILES
Search the current working directory for all PROPERTY-*.md files:
ls -t PROPERTY-*.md 2>/dev/null
Primary data sources (check for all of these):
| File Pattern | Data It Contains | PDF Section |
|---|
PROPERTY-ANALYSIS-*.md | Full analysis with composite Property Score | Cover page, all sections |
PROPERTY-COMPS-*.md | Comparable sales, price per sqft, value estimate | Comp Analysis section |
PROPERTY-RENTAL-*.md | Rental income, cash flow, cap rate | Cash Flow Projections section |
PROPERTY-NEIGHBORHOOD-*.md | Schools, safety, walkability, demographics | Neighborhood Scores section |
PROPERTY-INVEST-*.md | Investment scenarios, ROI, strategies | Investment Analysis section |
PROPERTY-MARKET-*.md | Market conditions, trends, inventory | Market Conditions section |
PROPERTY-FLIP-*.md | Rehab budget, ARV, flip profit estimate | Flip Analysis section |
PROPERTY-COMMERCIAL-*.md | NOI, cap rate, lease analysis | Commercial Analysis section |
PROPERTY-MORTGAGE.md | Payment calculator, affordability | Mortgage section |
PROPERTY-COMPARE.md | Side-by-side comparison | Comparison section |
PROPERTY-LISTING-*.md | MLS listing description | Listing section |
PROPERTY-SCREEN-*.md | Screener results | Screening section |
Find the most recent version of each:
ls -t PROPERTY-ANALYSIS-*.md 2>/dev/null | head -1
ls -t PROPERTY-COMPS-*.md 2>/dev/null | head -1
ls -t PROPERTY-RENTAL-*.md 2>/dev/null | head -1
ls -t PROPERTY-NEIGHBORHOOD-*.md 2>/dev/null | head -1
ls -t PROPERTY-INVEST-*.md 2>/dev/null | head -1
ls -t PROPERTY-MARKET-*.md 2>/dev/null | head -1
If no previous data exists:
- Recommend the user run
/realestate analyze <address> first for the best results
- If the user insists, ask for the property address and run a quick data collection using WebSearch to build the data structure from scratch
- At minimum, run the equivalent of
/realestate quick <address> to populate basic scores
STEP 3: EXTRACT DATA FROM ANALYSIS FILES
Read each found file and extract the key data points into a structured format:
From PROPERTY-ANALYSIS-*.md (primary source):
- Property address
- Property type (SFR, condo, multi-family, etc.)
- Listing price
- Beds / Baths / Square footage / Lot size / Year built
- Composite Property Score (0-100)
- Property Grade (A+ through F)
- Signal (Strong Buy through Avoid)
- Category scores: Value & Comps, Income Potential, Neighborhood, Investment, Market
- Key findings (bulleted list)
- Risk factors
- Recommendation summary
From PROPERTY-COMPS-*.md:
- Comparable sales list (address, price, sqft, beds/baths, distance, sale date)
- Estimated market value
- Price per square foot vs comps
- Over/under priced assessment
From PROPERTY-RENTAL-*.md:
- Estimated monthly rent
- Net monthly cash flow
- Cap rate
- Cash-on-cash return
- Gross rent multiplier
- Expense breakdown
- Vacancy assumption
From PROPERTY-NEIGHBORHOOD-*.md:
- School ratings (elementary, middle, high)
- Walk Score / Transit Score / Bike Score
- Safety rating
- Demographics summary
- Growth outlook
From PROPERTY-INVEST-*.md:
- Best investment strategy
- Projected ROI (1yr, 3yr, 5yr)
- Risk level
- Value-add opportunity description
- Exit strategy options
From PROPERTY-MARKET-*.md:
- Market type (buyer/seller/balanced)
- Median home price
- Days on market (average)
- Inventory months
- Price trend (YoY)
- Economic drivers
STEP 4: BUILD THE JSON DATA STRUCTURE
Assemble all extracted data into a structured JSON payload for the PDF generator:
{
"property_address": "123 Main Street, City, ST 12345",
"report_date": "April 29, 2026",
"property_type": "Single Family Residence",
"listing_price": 425000,
"beds": 3,
"baths": 2,
"sqft": 1850,
"lot_size": "7,200 sf",
"year_built": 2005,
"property_score": 76,
"grade": "A",
"signal": "Buy",
"categories": {
"Value & Comps": {
"score": 78
STEP 5: GENERATE THE PDF
Run the PDF generation script:
python3 ~/.claude/skills/realestate/scripts/generate_realestate_pdf.py
If the script does not exist, generate the PDF inline using Python and ReportLab. The inline script must produce a PDF with the following sections:
PDF SECTIONS AND LAYOUT
Page 1: Cover Page
- Report title: "Property Analysis Report"
- Property address (large, centered)
- Property Score gauge (circular, color-coded: green 70+, yellow 40-69, red 0-39)
- Grade and Signal displayed prominently
- Report date
- Disclaimer footer
Page 2: Property Overview
- Property details table (price, beds, baths, sqft, lot, year, type)
- Property photo placeholder or description
- Executive summary (2-4 sentences)
- Key findings list (bulleted, top 5)
Page 3: Comparable Sales Analysis
- Comp table: address, price, $/sqft, beds/baths, distance, sale date
- Estimated value vs listing price
- Price per sqft comparison bar chart
- Over/under priced assessment with percentage
Page 4: Cash Flow Projections
- Rental income estimate
- Monthly expense breakdown table (mortgage, taxes, insurance, vacancy, maintenance, management)
- Net monthly cash flow (highlighted, green if positive, red if negative)
- Key return metrics: Cap Rate, Cash-on-Cash, GRM
- 5-year cash flow projection table
Page 5: Neighborhood Scorecard
- School ratings (elementary, middle, high) with bar visualization
- Walk Score / Transit Score / Bike Score gauges
- Safety rating
- Demographics summary
- Growth outlook
- Amenities nearby
Page 6: Investment Analysis
- Category scores bar chart (all 5 categories)
- Best strategy recommendation
- Projected ROI table (1yr, 3yr, 5yr)
- Risk level assessment
- Value-add opportunity description
- Exit strategy options
Page 7: Market Conditions
- Market type indicator (buyer/seller/balanced)
- Median price and trend
- Days on market and inventory
- Economic drivers
- Price trend chart or table
- Supply/demand assessment
Page 8: Recommendation & Next Steps
- Overall recommendation (highlighted)
- Signal with explanation
- Key action items
- Suggested next steps
- Full disclaimer
PDF STYLING
| Element | Style |
|---|
| Colors | Navy (#1B2A4A) headers, dark gray (#333) body, green (#2E7D32) positive, red (#C62828) negative |
| Fonts | Helvetica-Bold for headers, Helvetica for body |
| Score gauges | Circular arc gauges with color gradient (red -> yellow -> green) |
| Tables | Alternating row colors (white/#F5F5F5), navy header row |
| Charts | Horizontal bar charts for category scores and comparisons |
| Footer | Page numbers, disclaimer, generation date |
| Margins | 50pt top, 40pt sides, 50pt bottom |
STEP 6: VERIFY AND DELIVER
After PDF generation:
ls -la PROPERTY-REPORT.pdf
Confirm the file was created and report:
- File name and location
- File size
- Number of pages
- Which data sources were included (list the PROPERTY-*.md files used)
- Any data gaps (sections that had no source file — these will show "Data not available" in the PDF)
OUTPUT SPECIFICATIONS
| Spec | Value |
|---|
| File name | PROPERTY-REPORT.pdf (or PROPERTY-REPORT-[ADDRESS].pdf if address specified) |
| Page size | Letter (8.5" x 11") |
| Orientation | Portrait |
| Pages | 6-10 depending on available data |
| File size | Typically 200KB - 1MB |
| Python dependency | ReportLab (pip install reportlab if not installed) |
RULES
- Professional quality — The PDF must look like it came from a real estate analytics firm, not a quick printout
- Data-driven — Every number in the PDF must come from the analysis files or live research; never fabricate data
- Conservative estimates — Use the same conservative projections from the analysis files
- Complete disclaimer — Full disclaimer must appear on the cover page and the last page
- Graceful degradation — If some analysis files are missing, generate the PDF with available data and mark missing sections as "Not analyzed — run /realestate [command] to add this data"
- Install dependencies — If ReportLab is not installed, install it automatically:
pip install reportlab
- Overwrite safely — If PROPERTY-REPORT.pdf already exists, overwrite it (the latest data wins)
- Color-coded scores — All scores must be color-coded: green (70+), yellow (40-69), red (0-39)
ERROR HANDLING
- If ReportLab is not installed, run
pip install reportlab and retry
- If no PROPERTY-*.md files exist, prompt the user to run
/realestate analyze <address> first
- If the Python script fails, capture the error message and display it to the user with troubleshooting steps
- If only partial data is available, generate a partial report and clearly mark which sections are incomplete
- If the PDF file cannot be written (permission error), suggest an alternative output directory
DEPENDENCY INSTALLATION
If ReportLab is not available, install it:
pip install reportlab 2>/dev/null || pip3 install reportlab 2>/dev/null
If installation fails, provide manual instructions:
To install the PDF generation dependency:
pip install reportlab
If using a virtual environment:
python3 -m venv venv && source venv/bin/activate && pip install reportlab
WHEN TO RECOMMEND PDF vs MARKDOWN
| Situation | Recommend |
|---|
| Client presentation or email attachment | PDF |
| Lender or partner due diligence package | PDF |
| Quick internal reference | Markdown |
| Iterative editing and analysis | Markdown |
| Board or investor meeting | PDF |
| Personal property shopping | Markdown |
| Sales collateral for real estate agent | PDF |
Always suggest: "Your analysis files are saved as Markdown for easy reference. Run /realestate report-pdf anytime to generate a polished PDF version for clients or presentations."
DATA QUALITY FLAGS
When compiling the PDF, flag data quality issues:
| Flag | Condition | Display In PDF |
|---|
| High Confidence | All 5 analysis agents ran, data is fresh | Green checkmark |
| Moderate Confidence | 3-4 agents ran, or data is 7+ days old | Yellow warning |
| Low Confidence | Only 1-2 agents ran, or significant data gaps | Red flag with note |
| Stale Data | Analysis files are 30+ days old | Warning banner: "Data may be outdated" |
MULTI-PROPERTY REPORTS
If the user has analyzed multiple properties (multiple sets of PROPERTY-*.md files), the PDF should:
- Detect all unique properties from file names
- Ask the user which property to include (or all)
- If "all", create a multi-property report with a comparison summary page
- Each property gets its own section with the standard layout
- Final page includes a side-by-side comparison table if 2+ properties are included
DISCLAIMER: For educational/research purposes only. Not financial or investment advice. All estimates are AI-generated approximations based on publicly available data. Always verify with licensed professionals and conduct your own due diligence before making any purchase or investment decisions.
