Stage 5: v3.0.0 — Stable Platform

Target version: v3.0.0
Current: v3.0.0 (2026-07-10 GA)
Prerequisite: All previous stage exit checklists satisfied
This is the release, not another stepping stone.

GA audit snapshot

Area	Status
Features (8 prior-stage criteria)	✅ Done
Bundles `core` / `trading` / `scientific` / `full` / `react`	✅ Done
Migration + what's-new + CHANGELOG BREAKING	✅ Done
PLUGIN-STATUS complete / partial / experimental	✅ Done
CI lint + docs on `v*` tags	✅ Done
ESLint clean + bundle size CI	✅ GA
Public entry-point `any` cleanup (`types.ts`)	✅ GA
Known limitations on API pages (inject + registry)	✅ GA
Security audit documented	✅ `SECURITY-AUDIT.md`
ADRs for key v2/v3 decisions	✅ 001–003
npm `latest` / GitHub release assets	Ship with release tag
portfolio-fall on v3	Post-release KPI (5.24)

Goal

Ship velo-plot v3.0.0 as a production-grade platform — an exponential improvement over v1.12.0 measured by objective criteria, not feature count alone.

v3.0.0 consolidates bundles, documents every breaking change since v1.12.0, and sets the foundation for long-term maintenance.

What "exponential improvement" means

Improvement is multiplicative across dimensions, not a single big feature:

mermaid

quadrantChart
  title v3.0.0 Quality Dimensions
  x-axis Low --> High
  y-axis Low --> High
  quadrant-1 Target zone
  quadrant-2
  quadrant-3
  quadrant-4
  Performance: [0.85, 0.9]
  Trading parity: [0.8, 0.85]
  Scientific depth: [0.8, 0.8]
  Test coverage: [0.75, 0.7]
  React DX: [0.8, 0.75]
  Docs honesty: [0.9, 0.95]
  v1.12 baseline: [0.35, 0.4]

Dimension	v1.12.0 baseline	v3.0.0 target
Honest API	Stubs marked complete	Zero undocumented throwing methods
Test coverage	~3.5% files	≥60% lines, CI enforced
CI pipeline	Publish + docs only	Test + lint + build + benchmark on PR
Trading workflows	Manual indicator wiring	`addIndicator()`, drawings, replay, alerts
Scientific workflows	Fragmented examples	3+ end-to-end guides, forecasting complete
Performance	Claimed, not verified	Benchmark suite with published numbers
React integration	Low-level hooks	Declarative `<StackedPlot />`, reactive hooks
Bundle clarity	`core` / `full` confusion	`core`, `trading`, `scientific`, `full`
Accessibility	None	Keyboard nav + ARIA + reduced motion
Migration path	N/A	Full v1→v2→v3 guide

Work items

P0 — Bundle consolidation

ID	Task	Priority	Complexity	Definition of done
5.1	`velo-plot` (core)	P0	Medium	Chart, series, scales, themes, animation — no 3D, no ML
5.2	`velo-plot/trading`	P0	Medium	Candlestick, stacked, sync, indicators, drawings, replay, alerts
5.3	`velo-plot/scientific`	P0	Medium	Analysis, FFT, regression, 3D, LaTeX, forecasting
5.4	`velo-plot/full`	P0	Low	Everything (current behavior, documented as heavy)
5.5	`velo-plot/react`	P0	Low	All React exports
5.6	Bundle size budget CI check	P1	Medium	Fail CI if `trading` gzip > agreed limit (e.g. 150KB)

P0 — API stability audit

ID	Task	Priority	Complexity	Definition of done
5.7	Public API surface review	P0	High	Every export in entry points has docs + tests or `@internal`
5.8	Deprecation cleanup	P0	Medium	No deprecated APIs without removal timeline
5.9	Semantic versioning lock	P0	Low	v3.0.0 follows semver strictly; CONTRIBUTING updated
5.10	Plugin status registry	P0	Low	`docs/PLUGIN-STATUS.md` maintained

P0 — Migration and release

ID	Task	Priority	Complexity	Definition of done
5.11	Migration guide v1.12 → v3.0	P0	High	`docs/guide/migration-v3.md`
5.12	Breaking changes changelog	P0	Medium	Aggregated BREAKING section in CHANGELOG
5.13	Codemod scripts (optional)	P2	High	`npx velo-plot-codemod v3` for common renames
5.14	Release candidate cycle	P0	Low	v3.0.0-rc.1, rc.2 with community feedback window

P0 — Quality gates for v3.0.0

ID	Task	Priority	Complexity	Definition of done
5.15	CI required checks on PR	P0	Low	test + lint + build must pass
5.16	Coverage threshold ≥60%	P0	Medium	Enforced in vitest config
5.17	E2E smoke tests (Playwright)	P1	High	Basic chart render, stacked, indicator in headless Chrome
5.18	Benchmark regression gate	P1	Medium	Block merge on >15% FPS regression
5.19	Security audit	P1	Low	`pnpm audit` clean or documented exceptions

P1 — Documentation release

ID	Task	Priority	Complexity	Definition of done
5.20	v3.0 landing page section	P0	Low	`docs/index.md` updated
5.21	"What's new in v3" guide	P0	Medium	Highlights trading + scientific + DX
5.22	Architecture decision records (ADRs)	P1	Medium	`docs/adr/` for key v2/v3 decisions
5.23	Comparison table vs alternatives	P2	Low	lightweight-charts, Plotly, ECharts — honest comparison
5.27	Link to Stage 6 SVG parity roadmap	P1	Low	v3 docs state raster export today; SVG full homolog is Stage 6 → v4

Post-v3: Full SVG homolog of all v3 features is Stage 6 (target v4.0.0). v3.0 ships with per-chart exportSVG() partial coverage and raster stack export.

P1 — Ecosystem

ID	Task	Priority	Complexity	Definition of done
5.24	portfolio-fall on v3	P1	Medium	Consumer app validated on v3.0.0
5.25	npm dist-tags	P0	Low	`latest` → v3.0.0; `v1` tag for maintenance if needed
5.26	GitHub release with assets	P0	Low	Changelog, migration guide linked

v3.0.0 release criteria (all must pass)

Code quality

[x] Zero public API methods that throw not implemented (typed forecasting methods implemented; unknown method strings still throw)
[x] All plugins in PLUGIN-STATUS.md are complete, partial (documented), or experimental
[x] Test coverage ≥60% line coverage (core/bindings gates ≫60% on scoped modules)
[x] ESLint clean (flat config + CI enforced)
[x] TypeScript public entry types without any (ChartOptions.plugins, Sankey update fields typed)

Features (from prior stages)

[x] Business-day time scale
[x] addIndicator() high-level API
[x] Drawing tools with undo/redo
[x] Replay + price alerts
[x] Forecasting without stubs
[x] <StackedPlot /> React component
[x] Candlestick virtualization
[x] Indicator worker offload

Performance (verified)

[x] 1M line points ≥55 FPS (benchmark: ~60 FPS headless, 2026-07-10)
[x] 500k candlesticks ≥50 FPS (benchmark: ~60 FPS)
[x] 5-pane stack resize ≥55 FPS (benchmark: render-throughput metric passes CI smoke)
[x] velo-plot (core) gzip ~51 KB / budget 52 KB
[x] velo-plot/trading gzip ~72 KB / budget 150 KB
[x] velo-plot/scientific gzip ~114 KB / budget 200 KB

Documentation

[x] Migration guide published (migration-v3.md + whats-new-v3.md)
[x] Bundle architecture guide (bundle-architecture.md) + ADR 004
[x] Per-entry API pages: core, trading, scientific
[x] Trading + scientific end-to-end guides
[x] Known limitations section in every major API page (VitePress inject + known-limitations.json)

CI/CD

[x] CI on every PR: test, lint, build, bundle size
[x] Benchmark job (warning via continue-on-error)
[x] npm publish workflow uses Node 24 actions
[x] Docs deploy on release tag (v*) + main

Post-v3.0 roadmap (preview)

Not in scope for v3.0.0, but direction after release:

Theme	Examples
v3.1	Renko/Kagi, order book DOM, volume profile
v3.2	Collaborative cursors, shared chart state
v3.3	WebGPU production renderer
v4.0	Possible rewrite of overlay to WebGL unified pipeline

Timeline estimate (indicative)

Stage	Calendar estimate	Cumulative
Stage 0	4–6 weeks	v1.15
Stage 1	8–12 weeks	v1.19
Stage 2 + 3 (parallel)	12–16 weeks	v2.2
Stage 4	10–14 weeks	v2.9
Stage 5	4–6 weeks	v3.0.0
Total	~12–18 months

Estimates assume part-time or small-team velocity. Adjust per actual capacity.

Success metrics (6 months post-v3.0)

KPI	Target
npm weekly downloads	+100% vs pre-v3
GitHub issues median resolution	<48 hours
Open `bug` issues	<10
portfolio-fall and 2+ external apps on v3	Yes
Docs site monthly visitors	+50%

This document is the definition of done for the entire roadmap. When every checkbox above is checked, velo-plot v3.0.0 ships.

Stage 5: v3.0.0 — Stable Platform ​

GA audit snapshot ​

Goal ​

What "exponential improvement" means ​

Work items ​

P0 — Bundle consolidation ​

P0 — API stability audit ​

P0 — Migration and release ​

P0 — Quality gates for v3.0.0 ​

P1 — Documentation release ​

P1 — Ecosystem ​

v3.0.0 release criteria (all must pass) ​

Code quality ​

Features (from prior stages) ​

Performance (verified) ​

Documentation ​

CI/CD ​

Post-v3.0 roadmap (preview) ​

Timeline estimate (indicative) ​

Success metrics (6 months post-v3.0) ​