Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.deckr-ai.com/llms.txt

Use this file to discover all available pages before exploring further.

Data architecture

Deckr routes data across three specialized database tiers based on data type:

SQL (PostgreSQL)

Structured financial data — ratios, loan terms, balance sheet figures, covenants. Queryable and reportable.

MongoDB

Long-form document text, news articles, court filings, and business reviews. Schema-flexible storage for unstructured content.

Neo4j (Graph)

Entity relationships — ownership structures, corporate affiliates, collateral linkages, and principal networks.

Data sources

SourceTypeUsed for
Uploaded PDFsBorrower-providedFinancial statements, tax returns
SerpAPIExternal enrichmentNews and web research on borrower
OpenCorporatesExternal enrichmentCorporate registry and ownership data
CourtListenerExternal enrichmentPublic court filings and litigation history
Yelp FusionExternal enrichmentBusiness reviews and reputation signals

Enrichment pipeline

After document extraction, Deckr asynchronously enriches the deal with external data. This runs in the background and does not block the agent pipeline. Enrichment results are incorporated into the Industry Analysis and Guarantor agents.

Integrations roadmap

The following integrations are planned as part of the Strategic Data Routing roadmap:
  • Market data — public equity tickers and trading data for publicly traded borrowers or competitors
  • Prediction markets — macro risk signals from Polymarket and Kalshi
  • Banking API — transaction-level cash flow data via Plaid or Finicity (with borrower OAuth consent)
(Coming Soon)