# CRE Acquisition Pre-review

> A local-first pre-review tool for a single US multifamily acquisition package.

## What this is

CRE Acquisition Pre-review extracts deal fields, calculates screening metrics when source data supports them, and links review findings to document evidence.

## Accepted inputs

- Offering Memorandum
- Rent roll
- T12 operating statement
- Lease summary
- Loan term sheet
- Formats: PDF, XLSX, CSV, and DOCX

## Reported outputs

- NOI, cap rate, DSCR, physical and leased occupancy, collected-rent ratio, and rent PSF when supported by source documents
- Missing documents, data conflicts, rent roll and expense anomalies, lease risks, debt constraints, and source evidence
- Partial, missing, conflicting, unsupported, and unverified states remain visible

## Public explanations

- Methodology: https://cre-due-diligence-ai.pages.dev/methodology/
- FAQ: https://cre-due-diligence-ai.pages.dev/faq/
- Data use: https://cre-due-diligence-ai.pages.dev/data-use/

## Short factual answers

- This is a pre-review for one ordinary US multifamily acquisition package, not a full CRE platform.
- NOI uses supported T12 income less property operating expenses; Cap Rate uses specified NOI divided by an asking or entered purchase price; screening DSCR uses trailing actual NOI divided by specified 12-month debt service.
- Missing or conflicting values remain visible and can block dependent metrics. The system does not silently fill missing values with zero.
- The public parser returns a transient parse summary. It does not provide valuation, investment, legal, lending, or buy/no-buy advice.

## Scope boundaries

This V0.1 is an acquisition pre-review for one ordinary US multifamily asset. It does not provide valuation, investment, legal, lending, or buy/no-buy advice. PSA, title, survey, CRM, marketplace, team collaboration, and external market-rent estimates are outside scope.

## Public demo

- Homepage: https://cre-due-diligence-ai.pages.dev/
- The online demo sends selected files to Cloudflare for one transient parse request and does not store them.
- Use sample or non-confidential files in the online demo.
- The complete Canonical CRE review, calculations, findings, and report export remain available in the local prototype.