Cross-Cutting Tools

Document to Database

Executable orchestrator that turns tokenized/extracted CRE document content (rent rolls, T-12s, operating statements, Prose Frontier narrative artifacts) into validated, typed, auditable, target-model-ready database payloads.

Download the CRE Skills Plugin

Latest release, portable bundle (signed). Review the SKILL.md files before installing into your agent.

human gateReview recommendeddataNo personal datatypeorchestrator
What it does

Takes tokenized CRE document content (rent rolls, T-12s, operating statements) and produces validated, source-cited, warehouse-ready payloads. Every record carries a full provenance trail and unmapped charges route to a human-review queue rather than being guessed.

Why it matters

Deal teams assemble data rooms with rent rolls in one format, T-12s in another, and no clean path from extracted documents to a model-ready database. Numbers land in spreadsheets that can't be traced back to a source cell, charges get collapsed into a single rent figure, and recoveries never tie to the T-12 because nobody normalized the charge schedule to begin with.

How it's done today

An analyst copies figures from a PDF rent roll into a spreadsheet, maps charge codes by hand, and reconciles the resulting totals against the operating statement using a manual tie-out column. The mapping logic lives in that analyst's head, PII fields sit unredacted in shared workbooks, and there is no grade or confidence score to tell the next person how much to trust the numbers.

When to use it

Reach for it

Use when extracted or tokenized data-room documents must become typed, validated, provenance-stamped records before feeding an underwriting model, a warehouse, or a reconciliation run. Also use when a mixed data room (rent rolls plus operating statements) must be classified, normalized, and graded as one ingestion.

Not the right tool

Not the right entry point for a single rent roll or T-12; use rent-roll-to-database or t12-to-database directly instead. If the rent roll and T-12 are already normalized and you only need to tie them out, use rent-roll-t12-tieout. To extract tokens from a raw document, use document-to-data-room-extractor upstream.

What it needs and produces

Inputs

  • OM
  • Rent Roll
  • T-12

Outputs

  • Workflow plan
Example use case

An acquisitions team receives a data room with a 120-unit rent roll, a partial-year T-12, and a broker OM. The skill classifies each document, pseudonymizes tenant names, maps charge codes against the canonical chart of accounts (flagging two unrecognized CAM codes for human review), validates the charge schedule against the operating statement, grades the ingestion at B/88 (merge-ready, not production-ready), and outputs a load plan for the normalized relational profile.

Compatible agents

Agent personas that pair well with this skill

Works with
Limitations

The skill carries contractual in-place income from the rent roll and recognized accrual from the T-12. It does not compute collected cash, delinquency, or bad debt. Charge and account mapping is deterministic: a novel code with no recognized alias lands in the human-review queue rather than being inferred. Escalations are captured but not stepped forward; that belongs to the downstream underwriting model.