ehoandClaude Opus 5
664a43bf2d
Runs the multi-agent refactoring-backlog pipeline in docs/project/ refactor-backlog-setup/ up to and including three of the seven Phase 1 agents. 00-baseline.md establishes the metrics every later agent must cite, using only tooling already in the repo (vitest lcov, coverlet cobertura, ESLint's core `complexity` rule at threshold 0 for a full distribution, depcruise --metrics). Duplication and C# complexity had no tooling, so tools/baseline-scan.mjs adds a deterministic ~200-line text scan rather than a new dependency; the approximations are labelled as such. Headline: FE 75.1% line coverage but only over the 98 of 220 source files a spec loads; BE 97.6% line / 79.6% branch; 0 layering violations; 7.1% duplication; 25 of 2085 TS functions over CC 10. Then 02-testability, 04-cqrs-light and 06-adr-conformance (27 findings). 01/03/05 were skipped deliberately — the baseline shows little for them to find; 07 (BIO2) and 08 (consolidation) are still open. Each agent corrected a baseline observation of mine, and in every case the error was in something derived rather than measured: - BL-007 counted ~13 adapter "mutations" from the `runSubmit` helper name; 5 of those call sites are reads. It also missed 3 real mutations that reach the raw ApiClient and never return a Result. - BL-002 diagnosed the 100%-duplicated auth folders as ADR-0002's divergence prediction failing. It never had a chance to fail: §3's `Principal` union was never built. - BL-004 named libs/shared/domain and libs/beheer/contracts as coverage gaps; both are pure type declarations where 0% is unimprovable. All three corrections are recorded inline in 00-baseline.md §10, so agent 08 does not inherit the bad numbers. .prettierignore excludes the agent prompt directories — reflowing their markdown would edit the prompt text itself. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Documentation
Docs are split by kind, and kept out of each other's way:
reference/— information. How the system works and why: architecture, decisions (ADRs), the FP/TEA/atomic learning guide, accessibility and UX reference. Stable knowledge, not tied to a sprint.project/— administration. Planning and tracking: the work-package backlog, product requirements (PRDs), and the (superseded) roadmap. This is the moving, process-facing material.
Teaching material that is best read next to the components lives in Storybook, not
here — see the Foundations section (libs/shared/docs/*.mdx, run npm run storybook).
The reference/ docs are the long-form source; the Foundations pages are the condensed,
cross-linked curriculum.
Starting out? Foundations → Learning Path (libs/shared/docs/learning-path.mdx) is a
paced, hands-on three-day route through the codebase; Foundations → Overview
(overview.mdx) is the map of every idea, cross-linked.
reference/ — information
| Doc | What it is |
|---|---|
| architecture/ARCHITECTURE.md | The architecture walkthrough: contexts/layers, state management, parse-don't-validate, the feature recipe, the .NET backend seam. |
| architecture/0001-bff-lite-decision-dtos.md | ADR — BFF-lite endpoints + decision DTOs (backend decides, FE renders). |
| architecture/0002-user-groups-and-bounded-contexts.md | ADR — user groups as actors; identity vs authorization. |
| architecture/0003-cibg-huisstijl.md | ADR — adopt CIBG Huisstijl (vendored Bootstrap 5.2) + the token bridge. |
| architecture/0004-stamdata-as-code.md | ADR — business-tunable reference data as typed, compile-time-validated config (not a production DB). |
| architecture/0005-openzaak-behind-bff.md | ADR — connect to OpenZaak (ZGW APIs) behind the BFF via a config-gated data-source seam; the FE never changes. |
| architecture/0006-test-data-builders.md | ADR — build test data through the production door: type-state builders, reducer replay, and which fixture idiom fits which test. |
| openzaak-integration.md | How the BFF sources cases from OpenZaak (the IZaakSource seam + ZGW client), and how to add the next slice. |
| ../backend/openzaak/README.md | Docker harness for running OpenZaak locally: bring-up, integration test, notifications, teardown. |
| stamdata.md | How stamdata (config-as-code reference data) is laid out, how to add a table with zero UI code, and why coupling stays low. |
| audit-log.md | How the data-minimised authz/PII-reveal audit trail is built, how to audit a new action, and the one-producer-hub coupling. |
| feature-flags.md | How runtime feature flags work (catalog-as-code + runtime state), how to add one, and the hand-wired gating coupling to watch. |
| scaffolding.md | How code generation & scaffolding work: plop generators (gen:value-object/gen:form-machine), the NSwag client (gen:api), showcase snippets, and the skill recipes. |
| roles-and-access.md | The roles/actors + capability model: who can do what, how to switch roles in dev, and what each unlocks. |
| architecture/dependencies.md | Bounded-context + atomic-layer boundaries: the allowed-import rules, how they're enforced (dep:check) and visualized (dep:graph). |
| architecture/dependency-graph.md | Generated mermaid graph of contexts × layers (regenerate with npm run dep:graph). |
| fp-tea-atomic-design.md | Long-form learning guide: FP + The Elm Architecture + atomic design. |
| wcag-checklist.md | Manual WCAG checks automation can't catch (tab order, focus traps, reflow). |
| ui-ux-audit.md | Early UI/UX audit against NL Design System (predates ADR-0003 — read in that light). |
project/ — administration
| Doc | What it is |
|---|---|
| backlog/README.md | The work-package backlog index — the live tracker, with the session protocol. |
| prd/0001-mijn-aanvragen-en-wizardstatus.md | PRD — "Mijn aanvragen": running wizards, application status, document preview. |
| prd/0002-attribute-based-access-control.md | PRD — attribute-based access control in the UI. |
| prd/0003-brief-v2-demo-script.md | Demo script — Brief v2 scenarios mapped to a URL + click path (WP-28). |
| SHOWCASE-ROADMAP.md | Superseded roadmap (absorbed into project/backlog/) — kept for history. |