EA Toolchain

The Enterprise Architecture toolchain produces structured architecture catalogs, interactive HTML diagrams, and per-client portals. It spans five stages from raw capture to published delivery.

Pipeline Overview

1
Capture
/toolkit:note-create
Raw notes from interviews, documents, or code
2
Extract
/ea:leanix-catalog-extract
Source → Objects + Relations + CO CSVs
3
Validate
Manual / Review Panel
CSV correction loop — human validation gate
4
Seed D1
npm run seed:<client>
Validated CSV → Cloudflare D1 database
5
Publish
npm run deploy:<client>
D1 → Cloudflare Pages site per client
6
Deliver
wrangler d1 export
Portable .sqlite + CSV bundle to client

Steps 1–3 are iterative. The D1 seed (step 4) only happens after validation is complete. The delivery bundle (step 6) is the contractual deliverable — it runs independently of the factory.

Meta-Model Reference

18 object type prefixes (CAP, APP, ITC…), 12 relationship types, TOGAF layers, hierarchy rules, and Macroscope deliverable mapping.

View Meta-Model →

Data Model

CSV schemas (objects, relations, CO), D1 database structure, customers.json config, and field definitions.

View Data Model →

Publishing Process

Step-by-step runbook: CSV validation → D1 seed → Cloudflare Pages deploy → JCT portal publish → delivery bundle generation.

View Runbook →

Skill Reference

All EA-related skills: catalog extraction, diagram generation, publishing, review panel — with usage, inputs, outputs, and pipeline position.

View Skills →

Governing Decisions

TFD-0019
EA Catalog Storage & Delivery Model
Per-customer D1 database, CSV-validated lifecycle, portable delivery bundle.
TFD-0017
JCT Portail Client Architecture
Cloudflare Pages + Access, one repo per client, template-driven onboarding.

Active Clients

Client D1 DB DAEs / Catalogs Portal
Transgesco transgesco DAE-0007 → TES + BCM transgesco.jacksoncreektech.ca
STM stm DAE-0001 → VTB, DAE-0002 → PIS stm.jacksoncreektech.ca