IDX data foundation · An ArziTech portfolio

IDX data that shows its sources and its gaps.

db-stock organizes prices, filings, profiles, news, and corporate actions into a traceable historical foundation—while keeping completeness, freshness, and conflicts visible.

Static snapshot · not a real-time feed · not an investment rating

LineageLedgerStatic interface example
BBCA · quarterly filing2025 Q4
RevenueIDR 106.4T
Company filingStored
Net incomeIDR 54.8T
Comparison sourceReview
Free cash flow
Not availableGap visible
source IDissuer-filingingestion run2026-08-02
Catalog snapshot

Broad coverage without pretending it is uniform.

The catalog count establishes scale. It does not promise equal depth, freshness, or quality for every ticker.

957tickers in the repository catalog snapshot

Snapshot dated August 2, 2026. Coverage varies by ticker and dataset; this is not a real-time feed.

Historical prices

Daily OHLCV with different date ranges across issuers.

Financial statements

Quarterly and annual periods with missing fields kept visible.

Key stats

Recomputed metrics and source snapshots kept intentionally distinct.

Company profiles

Identity, management, shareholders, and related entities.

News

Dated article archive with explicit source and freshness context.

Corporate actions

Dividends, splits, rights, bonus, meetings, and warrants where available.

Index context

Membership and index history as context, never as a signal.

Data lineage

Every value travels with its context.

The pipeline does not stop when a number is stored. Source, ingestion run, completeness audit, and verification results remain part of how the data is read.

  1. 01

    Source

    Company documents, market feeds, aggregators, and news archives.

  2. 02

    Ingestion

    Runs are recorded; payloads, periods, and units are normalized.

  3. 03

    Stored dataset

    PostgreSQL keeps observations without hiding null values.

  4. 04

    Audit

    Missing, stale, suspect, and conflicted states stay readable.

  5. 05

    Read surface

    Inspector, analysis UI, read-only API, and private agent integration.

Product surfaces

One foundation, several ways to read it.

This landing page only describes those surfaces. It does not connect to the private backend or carry any credential.

Completeness

Data inspector

Trace what exists, what is absent, and when each dataset was last updated.

Research

Analysis UI

Compare series, fundamentals, and trust context without erasing data limits.

Programmatic

Read-only API

An internal read contract; it is not exposed by this public landing page.

Authenticated

Private MCP integration

Agent access remains private, authenticated, scope-limited, and audited.

Transparency as a feature

Uncertainty is not polished away to look convincing.

db-stock treats data state as product information, not fine print. Readers can see the limits before drawing a conclusion.

Completeness measures coverage under project rules—not issuer quality, investment merit, or a recommendation.

Provenance records origin and ingestion path. It is not proof that a source was independently verified.

Missing

A value or period is not available.

Stale

Data is outside its freshness policy.

Suspect

A comparison needs human review.

Conflicted

Sources differ materially.

Built by ArziTech

An engineering portfolio about accountable data.

db-stock grew from a simple need: read Indonesian market history without losing its context. The project connects ingestion pipelines, provenance, audits, cross-source verification, automation, APIs, interfaces, and agent integration as one inspectable system.

Technical foundation
PostgreSQL 17TypeScriptNestJSReactn8nPrivate MCP
The goal is not to make every number look finished. It is to make each number's state clear enough to act on.
Future access

Follow along when a public surface is ready.

Register interest in future access or project updates. For now, this form only demonstrates the interface flow.

Preview form — your email is not stored or sent.

This prototype makes no network request and uses no cookie, localStorage, or database.