VynFi 1.x — DataSynth 5.5 Lands: Multi-Period Audit Data + Audit Methodology Library

DataSynth 5.5.0 published on crates.io two weeks ago. As of today VynFi's production engine, all three ACR images (CPU, LLM, GPU), and the dashboard wizard run against it end-to-end. This is Wave 1 of a four-wave 5.x adoption — the catch-up plus the customer-facing surfaces that depend only on the new binary.

Multi-period chain — full-year audit data without manual carryover

Audit firms have been asking for this since we shipped multi-period sessions in early 4.x. The old session model gave you N independent jobs that you had to thread together yourself: download Q1, extract closing trial balances, reformat as opening balances, push into Q2. By Q3 you've spent more time on file plumbing than on actual audit testing.

DataSynth 5.3 fixed this in the engine. The 5.5 release wires it through the API. Set `multiPeriod.enabled = true` and provide a `periodSpecs[]` array with 2–12 strictly-increasing fiscal periods, and the chain runs as a single job: closing TB of period N becomes opening TB of period N+1, FX revaluation runs at each period boundary, retained-earnings rolls correctly, and document chains span quarter boundaries.

Pricing — pay-per-period, no chain premium

We deliberately chose not to charge a multi-period premium. A 4-quarter chain costs exactly 4× what a single-period job would cost: 50,000 rows on a manufacturing config (1.5x sector multiplier) is 75,000 credits per quarter, 300,000 credits for the chain. The convenience is in the carryover plumbing, not the pricing — if you don't need temporal coherence, run four standalone jobs and pay the same total.

Multi-period is gated to the Team plan ($199/mo) and above. Free and Developer tiers receive a `tier_restricted` validation error pointing at the upgrade path. The Generate wizard now exposes the toggle on the Scale step; Scale+ customers can override per-period parameters via raw YAML.

Audit Methodology Library — public, free, SEO-indexed

The new public catalog at /audit-methodology surfaces every embedded reference in the upstream `datasynth-audit-fsm` crate as a browseable, no-auth-required surface. 19 GET endpoints under `/v1/audit-methodology/*`, ~50 SEO-indexed portal pages, sitemap and llms.txt updated for AI crawler discovery.

• **Big4 methodology** — common ISA-derived spine plus four firm overlays (EY GAM, PwC Aura, KPMG Clara, Deloitte Omnia) and a cross-firm equivalence map.
• **Jurisdictional overlays** — 7 regulatory regimes (PCAOB US, EU CSRD, UK FRC, ASIC AU, JFSA JP, ACRA SG, HKICPA HK) with 39 procedures total.
• **Methodology blueprints** — ISA 600 (group audit) and CSRD limited-assurance.
• **KYC blueprints** — 6 workflows: private banking onboarding, correspondent banking, crypto-CASP, periodic review, SAR escalation, sanctions remediation.
• **Banking forms** — 7 forms with 228 fields total: MROS SAR, four UBS reconstruction forms, Wolfsberg CBDDQ v1.4, Wolfsberg FCCQ v1.2. Plus a cross-form canonical-terms evidence index.
• **Engagement scenarios** — 15 deterministic scenarios with expected outcomes, from clean unqualified opinions to going-concern uncertainty to IFRS 8 segment complexity.
• **Reference content** — RMM 12-factor taxonomy, L4 graph schema (engagement → component → account → assertion → control → procedure → evidence), ISA 230 working-paper structure.

The catalog is read-only. It surfaces blueprints, overlays, forms, scenarios, taxonomies, and schemas — but does not execute any of them. Active surfaces (FSM execution, RMM Bayesian scoring, working-paper Merkle bundle generation, L4 graph queries) ship in Wave 4.

v5.1 polish — `accountNameOverrides` and SAP CEPC

Two smaller config-level additions for Team+ and Scale+ customers. `accountNameOverrides` is a dictionary that lets you remap the engine's default account names to your firm's chart-of-accounts conventions — useful when prepping data for a specific client engagement. The schema-level support is in 5.5; the runtime wiring (engine-side surface for the override map) lands in a future DataSynth point release. Until then the API accepts and tier-gates the override map but the engine ignores it.

SAP CEPC (profit-center master data) is now selectable in the SAP integration pack on Scale+. This pairs with BKPF/BSEG/ACDOCA for customers running profit-center-aware S/4HANA tests.

How to try it

Multi-period: in the Generate wizard, set Scale → Multi-period chain → 4 quarters → Manufacturing. Or via the API:

{
  "preset": "manufacturing_mid",
  "rows": { "journal_entries": 50000 },
  "multiPeriod": {
    "enabled": true,
    "periodSpecs": [
      { "fiscalPeriod": "2025-Q1", "periodMonths": 3 },
      { "fiscalPeriod": "2025-Q2", "periodMonths": 3 },
      { "fiscalPeriod": "2025-Q3", "periodMonths": 3 },
      { "fiscalPeriod": "2025-Q4", "periodMonths": 3 }
    ]
  }
}

Audit Library: https://vynfi.com/audit-methodology — no signup, no auth. Want machine-readable data? Hit any endpoint under `/v1/audit-methodology/*` from anywhere; no API key required.

Roadmap — Waves 2 to 4

DataSynth 5.5 also includes group-audit (5.0), acquisition accounting (5.2 + 5.4), and the active audit-methodology surface (5.5 FSM execution + RMM scoring). We deliberately deferred those to Waves 2–4 of the adoption to avoid bundling SKU and tenancy decisions with the catch-up release. Wave 2 (Group Audit MVP — `/v1/groups`, manifest/shard/aggregate, IFRS 10/IAS 21/IAS 28 baseline) starts when audit-firm signup signal from Wave 1 lets us calibrate the SKU model.

Full release notes are at the Changelog. The DS 5.5 adoption design lives at `docs/plans/2026-05-05-datasynth-5.5-adoption-design.md`. Questions: support@vynfi.com.