Callil Consolidation — Project Blueprint

Design Principles

Each run is independently verifiable — build, lint, tsc, tests, and VRT (where applicable) must pass at the end of every run.
Expose risks early — visual changes (theme migration) and complex new code (auto-publish) are scheduled before mechanical integration work.
Clear entry/exit criteria — each run has prerequisites, deliverables, and a verification gate. A run is not complete until its gate passes.
Incremental commits — each run produces one or more commits on the target branch. No run depends on uncommitted work from a prior run.

Playwright MCP for Visual Verification

Steps marked with [Playwright MCP] can use the Playwright MCP server to automate visual verification against the running Storybook dev server. This replaces manual spot-checks with reproducible browser-driven inspection:

Navigate to specific story URLs
Take screenshots for visual comparison
Interact with components (click, type, hover) to verify behavior
Compare against dev-witness stories side-by-side

Start Storybook (npm run dev in ux-prototype/) before running Playwright MCP checks.

Target Branch

jmpicnic/component-consolidation created off HEAD of jmpicnic/list-view-management-620 in the main clone (ux-prototype/).

Analysis Documents

All design decisions are recorded in analysis/:

Document	Governs
analysis/index.md	Master decisions and dispatch
analysis/primitives.md	Component organization, organism import mapping
analysis/styles.md	CSS tokens, globals, sonner/next-themes
analysis/dependencies.md	package.json changes, library build
analysis/grid-integration.md	Entity-data-grid evolution
analysis/test-coverage.md	Stories, tests, VRT, play function convention
analysis/canary-refactor.md	Follow-on adaptation

Run 0: Branch Setup

Risk: None Effort: Trivial

Entry Criteria

Main clone on jmpicnic/list-view-management-620 with clean working tree
Worktree at callil-consolidation-worktree on jmpicnic/callil-consolidation available for reference

Deliverables

Branch jmpicnic/component-consolidation created off HEAD of jmpicnic/list-view-management-620
Verified: npm ci && npm run lint && npx tsc --noEmit && npm run test pass on the new branch (baseline)

Exit Criteria

Green baseline on the new branch — all existing tests pass, no regressions from branch creation

Run 1: Foundation

Risk: High blast radius (many file moves and import rewrites), low per-file risk Effort: Medium Governs: analysis/primitives.md, analysis/styles.md

Intermediate gates are blocking. If an intermediate gate fails, stop and fix before proceeding to the next task. This isolates failures to the task that caused them. The final verification gate re-runs everything as a comprehensive check before committing.

This run establishes the new directory structure, design tokens, and import conventions that everything else depends on.

Entry Criteria

Run 0 complete (branch exists, baseline green)

Tasks

1a. Styles and Tokens

Update src/styles/tokens.css to Arda orange palette (worktree values), add new tokens (control heights, state ramp, font vars, destructive-foreground, dark mode)
Refactor src/styles/globals.css to @import './tokens.css' instead of redeclaring values
Replace remaining globals.css content with worktree version (Arda orange, Geist Mono font, touch media queries, sidebar-inset dark override)
Drop legacy sidebar CSS (.sidebar-menu-button-hover, [data-active='true'], .add-company-button)
Drop duplicate :root block

Intermediate gate (1a): npx tsc --noEmit + npm run test — token changes could break any component referencing CSS vars.

1b. Primitives Directory

Create src/components/canary/primitives/
Copy 13 stock shadcn files from worktree src/components/ui/ (collapsible, dropdown-menu, input, label, separator, sheet, sidebar, skeleton, table, tabs, textarea, toggle, tooltip)
Rewrite cn imports in primitives from @/utils to @/types/canary/utils
Do NOT copy sonner.tsx (dropped per styles.md decision)

Intermediate gate (1b): npx tsc --noEmit — verifies imports resolve. Tests won’t exercise primitives directly (no consumers yet).

1c. Custom Atoms

Copy custom components from worktree src/components/ui/ to src/components/canary/atoms/:
- avatar.tsx → atoms/avatar/
- badge.tsx → atoms/badge/ (merge with existing callil ArdaBadge wrapper)
- button.tsx → atoms/button/ (merge with existing callil ArdaButton wrapper)
- card.tsx → atoms/card/
- dialog.tsx → atoms/dialog/
- input-group.tsx → atoms/input-group/
Copy existing callil canary atoms from worktree: brand-logo, drawer, icon-button, icon-label, read-only-field, search-input (with their stories and tests)
Rewrite cn imports from @/utils to @/types/canary/utils
Rewrite @/components/ui/* imports to @/components/canary/primitives/*

Intermediate gate (1c): npx tsc --noEmit + npm run test — atoms have unit tests from the worktree; verifies they pass in the new location.

1d. Hook Migration

Copy src/hooks/use-mobile.ts from worktree to src/types/canary/hooks/use-mobile.ts
Update any imports referencing @/hooks/use-mobile to @/types/canary/hooks/use-mobile

Intermediate gate (1d): npx tsc --noEmit — small change, verify the import resolves.

1e. cn Import Rewrite

Rewrite 38+ files from import { cn } from '@/utils' to import { cn } from '@/types/canary/utils'
Verify src/utils.ts is not needed (if no other exports, do not copy it)

Intermediate gate (1e): npx tsc --noEmit + npm run lint + npm run test — full gate. This is the highest-risk task in Run 1 (broadest file change set). Any missed or malformed import shows up here. Lint catches import ordering. Run this gate before 1f so cn issues are debugged in isolation.

1f. Primitives Import Check Story

Add src/components/canary/primitives/primitives.stories.tsx — one integration story importing from several primitives and rendering a composed example

Intermediate gate (1f): npm run build-storybook — verifies Storybook can index and build with the new structure. [Playwright MCP] Navigate to the primitives import check story and verify components render.

Deliverables

New directory structure: canary/primitives/, updated canary/atoms/, types/canary/hooks/
Updated tokens.css and globals.css
All imports rewritten to canonical paths
Primitives import check story

Final Verification Gate (Run 1)

npm run lint — pass
npx tsc --noEmit — pass
npm run test — pass (all existing tests, no regressions)
npm run build-storybook — pass
VRT: generate new baselines for all new atom stories (badge, button, drawer, etc.)
[Playwright MCP] Spot-check new atom stories (badge, button, drawer, etc.) render correctly in Storybook

Exit Criteria

All files in canonical locations with canonical import paths
No references to @/components/ui/* or @/utils remain (except in vendored code)
Build green, Storybook builds

Run 2: Visual Foundation (Theme Migration)

Risk: Highest visual risk — diffs every grid story Effort: Small (code change is small, validation is thorough) Governs: analysis/grid-integration.md (decision #5)

Entry Criteria

Run 1 complete (styles, primitives, atoms in place)

Tasks

Retrofit DataGrid molecule (src/components/canary/molecules/data-grid/data-grid.tsx) from 'legacy' theme string to themeQuartz with design-system token mapping
Apply item-grid’s theme configuration: row height (48px), header height (36px), font (Geist Sans 14px, 13px headers, 600 weight), cell padding (12px), color token mapping, column border styling, wrapper border radius
Add ThemedGrid visual story to DataGrid molecule

Deliverables

DataGrid molecule using themeQuartz with Arda design tokens
ThemedGrid story

Verification Gate

npm run test — pass (existing DataGrid tests)
VRT: before/after comparison is the critical gate. Every grid story will diff — this includes DataGrid molecule stories, EntityDataGrid factory stories, EntityDataGridShim stories, and any other stories rendering an AG Grid. Expected variances: row heights, fonts, colors, borders all change from browser defaults to Arda tokens. Regenerate all affected baselines after review confirms visual correctness.
[Playwright MCP] Navigate to DataGrid and EntityDataGrid stories; screenshot before and after the theme change. Visually verify: row heights, header styling, font rendering, cell padding, color tokens applied correctly. Compare ThemedGrid story against item-grid’s existing appearance in the worktree Storybook for reference.

Exit Criteria

All grid stories render with Arda-themed appearance
VRT baselines regenerated and committed
No regressions in non-grid stories

Run 3: Organism Port

Risk: Medium (many files, but mechanical) Effort: Medium-Large Governs: analysis/primitives.md (organism import mapping section)

Move callil organisms and their supporting molecules into the main clone as-is (with import rewrites only, no behavioral changes). This validates that the foundation from Run 1 supports the organisms correctly.

Entry Criteria

Run 2 complete (themed DataGrid in place)

Tasks

3a. Supporting Molecules

Copy from worktree to main clone src/components/canary/molecules/:
- sidebar/ (sidebar-header, sidebar-nav, sidebar-nav-item, sidebar-nav-group, sidebar-user-menu) — with stories and tests
- item-grid/ (item-grid-columns, item-grid-fixtures, typeahead-cell-editor, select-cell-editor) — with stories and tests
- item-details/ (item-details-header, item-details-card-preview, action-toolbar, field-list) — with stories and tests
- overflow-toolbar/ — with stories
Rewrite all @/components/ui/* imports per the mapping in primitives.md
Rewrite @/utils → @/types/canary/utils

3b. Organisms

Copy from worktree to main clone src/components/canary/organisms/:
- sidebar/ (sidebar.tsx, sidebar-context.tsx, sidebar.stories.tsx, sidebar.test.tsx, sidebar.mdx, index.ts)
- app-header/ (all files)
- item-details/ (all files)
- item-grid/ (all files)
Rewrite all imports per mapping

3c. Assets

Copy any assets from the worktree (callil-consolidation-worktree/public/canary/) into the main clone (ux-prototype/public/canary/) that the organisms reference (brand logos, images), if not already present in the main clone
Note: all copy operations in Run 3 follow this same direction — source: worktree (callil branch, read-only reference) → target: main clone (component-consolidation branch, where edits happen)

Deliverables

All 4 callil organisms and their supporting molecules in the main clone
All imports rewritten to canonical canary paths
All organism stories render in Storybook

Verification Gate

npm run lint — pass
npx tsc --noEmit — pass
npm run test — pass (existing + new organism tests)
Storybook build succeeds (npm run build-storybook)
[Playwright MCP] Navigate to each organism’s stories in Storybook and verify rendering:
- Sidebar: Default, WithBadges, Composition (check nav items, dark theme, collapse/expand)
- AppHeader: Default, WithSearch (check action buttons, search input)
- ItemGrid: Default, WithSearch, Editable (check grid renders with data, search filters)
- ItemDetails: Default (check drawer opens, fields display, tabs work)

Exit Criteria

All organisms functional in Storybook with correct imports
No references to @/components/ui/* in any canary code

Run 4: Cell Editor Upgrade

Risk: Medium (replaces existing atom, could break grid stories) Effort: Small-Medium Governs: analysis/grid-integration.md (decision #2)

Entry Criteria

Run 3 complete (organisms ported, including callil’s select-cell-editor molecule)

Tasks

Promote callil’s SelectCellEditor from molecules/item-grid/ to atoms/grid/select/ as a first-class canary atom
Generalize options format to accept both SelectOption[] and Record<string, string>
Update the atom to follow canary conventions (StaticConfig/RuntimeConfig split, factory function createSelectCellEditor)
Add stories: Default, WithManyOptions, KeyboardNavigation, BothFormats
Add unit tests: highlight cycling, Enter/Escape, ARIA roles, checkmark rendering
Add play function with step DSL: open → Arrow Down 3x → Enter → verify
Update EnumCellEditor to delegate to SelectCellEditor (or replace entirely and update all references)
Delete deprecated EnumCellEditor if fully replaced

Deliverables

src/components/canary/atoms/grid/select/ with full story/test/play coverage
EnumCellEditor replaced or deprecated
All grid stories using enum editors still work

Verification Gate

npm run test — pass
New SelectCellEditor stories render correctly
Existing grid stories with enum columns render correctly (regression check)
VRT: capture SelectCellEditor popup baseline
[Playwright MCP] Navigate to SelectCellEditor stories; verify popup rendering, keyboard navigation (Arrow Down cycling, Enter to select, Escape to cancel), checkmark on selected item. Also check an existing enum grid story to confirm no regression.

Exit Criteria

Single canonical select/enum cell editor in canary atoms
All consumers updated

Run 5: Entity-data-grid Evolution

Risk: Highest complexity — core of the project Effort: Large Governs: analysis/grid-integration.md (decisions #1, #3, #4), analysis/test-coverage.md

Promote general-purpose capabilities from item-grid into entity-data-grid. Each sub-run adds one capability with its own stories and tests. Sub-runs are ordered by risk (highest first).

Entry Criteria

Run 4 complete (SelectCellEditor in place)

Sub-run 5a: Row Auto-Publish Lifecycle

Risk: Highest — new editing paradigm replacing cell-granular mode

Implement row-auto-publish as the entity-data-grid editing mode
Remove cell-granular onEntityUpdated callback
Add onRowPublish(rowId, changes, entity?) async callback
Implement pending changes accumulation per row
Implement 50ms debounce on row blur before publish
Add row visual states: idle → saving (.ag-row-saving) → success/error (.ag-row-error)
Implement imperative ref API: saveAll(), discardAll(), getDirtyRowIds()
Add stories: RowAutoPublish, RowAutoPublishError, SaveAllDrafts, DiscardAllDrafts (all with step DSL play functions)
Add unit tests for pending changes, publish lifecycle, ref API, CSS class injection
Gate: stories + tests + play functions pass; VRT for row visual states
[Playwright MCP] Navigate to RowAutoPublish story; edit a cell, edit another cell in same row, click a different row. Verify: saving row background appears briefly, then clears. Navigate to RowAutoPublishError; verify error row background persists. Check SaveAllDrafts/DiscardAllDrafts via toolbar button interactions.

Sub-run 5b: Actions Column

Risk: Medium — new InitConfig property

Add actionsColumn as InitConfig property (mount-time ColDef with optional actionCount)
Inject as pinned-right column with auto-width calculation
Add story: WithActionsColumn (with step DSL play function)
Add unit tests: column injection, width calc, action callbacks
Gate: story + test + play function pass
[Playwright MCP] Navigate to WithActionsColumn story; click action buttons in rows; verify visual pinned-right column appearance.

Risk: Medium — new UI elements in entity-data-grid

Add searchConfig to factory config (field list, placeholder)
Implement search bar with 150ms debounce
Implement filtered count display (“3 of 12 items” / “3 of 12 selected”)
Add stories: WithSearch, WithSearchAndSelection (with step DSL play functions)
Add unit tests: debounce timing, filter predicate, count updates
Gate: stories + tests + play functions pass
[Playwright MCP] Navigate to WithSearch story; type a search term; verify row count updates after debounce; verify count display text. Check WithSearchAndSelection: select rows, verify count switches to “X of Y selected”.

Sub-run 5d: Pagination Modes

Risk: Low-Medium — StaticConfig property, both modes already implemented elsewhere

Add pagination mode as StaticConfig (design-time) property
Support server-driven (PaginationData + callbacks) — existing behavior
Support client-side (AG Grid built-in pageSize) — from item-grid
Modes are mutually exclusive
Add story: ClientPagination; verify existing ServerPagination still works
Add unit tests: both modes produce correct grid config
Gate: stories + tests pass

Risk: Low — additive features

Add toolbar prop (ReactNode slot above grid)
Add autoHeight StaticConfig property (domLayout: 'autoHeight')
Add enableDragToScroll opt-in with useDragToScroll hook (~80 lines pointer-event logic)
Add stories: WithToolbar, AutoHeight, DragToScroll
Add unit tests: toolbar render, autoHeight pass-through
Gate: stories + tests pass

Deliverables (Run 5 overall)

Entity-data-grid with all promoted capabilities
10+ new stories with step DSL play functions
Comprehensive unit test coverage for each capability
VRT baselines for row visual states

Exit Criteria

Entity-data-grid API supports all capabilities needed for item-grid to use it as a base
All new stories render and pass
All existing entity-data-grid stories still pass (no regressions)

Run 6: Item-grid Integration

Risk: Integration risk — the core consolidation goal Effort: Medium-Large Governs: analysis/grid-integration.md (overall goal)

Refactor item-grid to use entity-data-grid as its base instead of building directly on the DataGrid molecule.

Entry Criteria

Run 5 complete (entity-data-grid has all needed capabilities)

Tasks

Evaluate whether item-grid can be a pure factory config call (createEntityDataGrid(itemGridConfig)) or needs a thin wrapper
Refactor item-grid implementation:
- Replace useItemGridEditing with entity-data-grid’s built-in auto-publish
- Replace inline search with entity-data-grid’s search/filter
- Replace inline drag-to-scroll with entity-data-grid’s built-in
- Replace inline toolbar with entity-data-grid’s toolbar slot
- Replace inline actions column with entity-data-grid’s InitConfig actions
- Keep item-domain-specific code: curated columns, typeahead lookups, nested field value getters/setters, custom renderers
Verify item-grid stories render identically to pre-refactor
Update item-grid tests

Deliverables

Refactored item-grid using entity-data-grid as base
Reduced item-grid code (domain config only, no framework reimplementation)
item-grid stories visually equivalent to pre-refactor

Verification Gate

npm run test — pass
Item-grid stories render correctly (VRT comparison against Run 3 baselines)
Entity-data-grid stories still pass (no regressions from refactoring)
Play functions in item-grid stories pass
[Playwright MCP] Navigate to all item-grid stories (Default, Empty, Loading, WithSelection, WithSearch, Editable, Paginated, Composition). Compare visually against the same stories rendered from the worktree Storybook to verify equivalence. Key checks: search filtering, editable cells with save lifecycle, selection count, pagination, actions column, drag-to-scroll.

Exit Criteria

Item-grid delegates to entity-data-grid for all general-purpose capabilities
No duplicated framework code between item-grid and entity-data-grid

Run 7: Package API

Risk: Low (mechanical, but affects published package) Effort: Small Governs: analysis/dependencies.md, analysis/index.md (barrel exports)

Entry Criteria

Run 6 complete (all components in final state)

Tasks

7a. Barrel Exports

Update src/canary.ts: union of both branches minus ArdaDetailField, plus all worktree additions (sidebar, app-header, item-grid, new atoms, new molecules)
Note: ArdaBadge is exported from both canary.ts and extras.ts — this collision is acceptable (different import paths). The canary version is canonical long-term; the extras version must not be used outside **/extras/** paths.
Verify src/extras.ts and src/index.ts unchanged (identical in both branches)

7b. Dependencies

Add radix-ui to production dependencies
Remove @radix-ui/react-tooltip (superseded)
Add to devDependencies: @reduxjs/toolkit, react-redux, redux-persist, sonner
Remove from devDependencies: next-themes, react-icons, shadcn, pako, qr-scanner, react-dropzone
Resolve @storybook/addon-links (drop if no linkTo usage confirmed)

7c. Cleanup

Delete src/components/canary/atoms/detail-field/ (replaced by ReadOnlyField)
Delete src/components/ui/ directory if still present (all contents moved)
Delete src/utils.ts if no remaining references
Delete src/hooks/use-mobile.ts if moved to types/canary/hooks/

Deliverables

Updated barrel exports
Updated package.json
No orphaned files

Verification Gate

npm run lint — pass
npx tsc --noEmit — pass
npm run test — pass
npm run build:lib — pass (library build produces dist/)
Verify dist/ contains expected entry points

Exit Criteria

Published package API reflects consolidated component library
Library builds successfully
No unused dependencies

Run 8: Acceptance

Risk: Low (verification, not implementation) Effort: Medium Governs: analysis/test-coverage.md (composition stories section)

Entry Criteria

Run 7 complete (package API finalized)

Tasks

8a. Composition Stories

Add items browse/view/edit stories in src/use-cases/reference/items/ (REF::ITM::0001, 0002, 0004)
Add suppliers browse/view stories in src/use-cases/reference/business-affiliates/ as canary variants (REF::BA::0001, 0002)
Add list view general behavior stories in src/use-cases/general-behaviors/list-views/ (GEN::LST::0002, 0003, 0004, 0006)
Add KitchenSink developer reference in entity-data-grid/
All play functions use step DSL

8b. VRT Baselines

Generate VRT baselines for all new composition pages
Generate VRT baselines for any components not yet captured
[Playwright MCP] Visual comparison of composition stories against dev-witness equivalents: navigate to each composition story and the corresponding dev-witness story side-by-side; screenshot both; verify layout, spacing, colors, and interaction patterns are recognizably similar

8c. Documentation

Ensure MDX docs exist for all components at every level:
- Primitives: one primitives.mdx describing the purpose of the primitives layer, what’s included, shadcn/ui origin, import convention, and regeneration instructions. No individual primitive descriptions needed.
- Atoms: MDX for each new atom (badge, button, avatar, card, dialog, drawer, input-group, icon-button, icon-label, read-only-field, search-input, SelectCellEditor)
- Molecules: MDX for new molecules (sidebar sub-components, action-toolbar, field-list, overflow-toolbar)
- Organisms: MDX for all organisms (sidebar, app-header, item-details — currently lacks MDX, item-grid, entity-data-grid, entity-data-grid-shim)
Update any stale MDX docs from the worktree that reference old import paths

Deliverables

6+ composition stories mapped to product use cases
Complete VRT baseline set
MDX documentation for all organisms

Verification Gate

npm run test — pass
npm run build-storybook — pass
npm run test:storybook — all play functions pass
VRT baselines committed
[Playwright MCP] Final walkthrough of all composition stories: navigate through items page (sidebar → grid → search → row click → drawer), suppliers page (grid → search → row click), KitchenSink (exercise all capabilities). Verify end-to-end interaction flow works.

Exit Criteria

Consolidated component library is proven to reproduce dev-witness page layouts using only canary components
Full test/story/VRT coverage per test-coverage.md metrics

Run 9: canary-refactor Adaptation (Follow-on)

Risk: Medium (may surface component gaps) Effort: Medium Governs: analysis/canary-refactor.md

Entry Criteria

Run 8 complete (acceptance passed)

Tasks

Adapt existing src/canary-refactor/ stories to use consolidated canary components
Determine if ItemDetailsPanel.tsx was absorbed into item-details organism or needs recreation
Determine if ItemTableAGGrid.tsx and columnPresets.tsx are still needed
Use divergences found during adaptation to drive fixes back into the component library

Deliverables

canary-refactor stories using consolidated components
List of component fixes/enhancements discovered during adaptation

Verification Gate

canary-refactor stories render and pass
Visual comparison against dev-witness originals

Exit Criteria

canary-refactor stories demonstrate visual/behavioral equivalence with vendored pages
All discovered issues resolved or tracked

Pre-Push Gate (All Runs)

Before any push to origin, the following must pass locally:

npm run lint
npx tsc --noEmit
npm run test
VRT if snapshots changed (npx playwright test --project=vrt)

Risk Summary

Risk	Run	Mitigation
Theme migration diffs every grid story	Run 2	Scheduled early; VRT before/after is explicit gate
Auto-publish is complex new code	Run 5a	First sub-run of capability promotion; isolated with its own tests
Import rewrite blast radius (38+ files)	Run 1	Mechanical with grep verification; build gate catches misses
SelectCellEditor replaces existing atom	Run 4	Existing enum stories serve as regression tests
Item-grid refactoring may need wrapper	Run 6	Entity-data-grid capabilities proven in Run 5 before refactoring begins
Barrel export changes affect published API	Run 7	Library build gate; deferred until all components finalized
canary-refactor may surface gaps	Run 9	Explicitly last; gaps drive fixes back into library

Dependency Graph

Run 0: Branch Setup
  └─► Run 1: Foundation (styles, primitives, atoms, imports)
        └─► Run 2: Visual Foundation (theme migration)
              └─► Run 3: Organism Port (sidebar, app-header, item-details, item-grid as-is)
                    └─► Run 4: Cell Editor Upgrade (SelectCellEditor)
                          └─► Run 5: Entity-data-grid Evolution
                          │     ├─ 5a: Auto-publish lifecycle
                          │     ├─ 5b: Actions column
                          │     ├─ 5c: Search/filter UI
                          │     ├─ 5d: Pagination modes
                          │     └─ 5e: Toolbar, auto-height, drag-to-scroll
                          │
                          └─► Run 6: Item-grid Integration
                                └─► Run 7: Package API
                                      └─► Run 8: Acceptance (composition stories)
                                            └─► Run 9: canary-refactor (follow-on)

All runs are sequential — each depends on the prior run’s exit criteria being met. Sub-runs within Run 5 are also sequential (5a → 5b → 5c → 5d → 5e) to allow incremental verification.

Callil Consolidation — Project Blueprint

Design Principles

Playwright MCP for Visual Verification

Target Branch

Analysis Documents

Run 0: Branch Setup

Entry Criteria

Deliverables

Exit Criteria

Run 1: Foundation

Entry Criteria

Tasks

1a. Styles and Tokens

1b. Primitives Directory

1c. Custom Atoms

1d. Hook Migration

1e. cn Import Rewrite

1f. Primitives Import Check Story

Deliverables

Final Verification Gate (Run 1)

Exit Criteria

Run 2: Visual Foundation (Theme Migration)

Entry Criteria

Tasks

Deliverables

Verification Gate

Exit Criteria

Run 3: Organism Port

Entry Criteria

Tasks

3a. Supporting Molecules

3b. Organisms

3c. Assets

Deliverables

Verification Gate

Exit Criteria

Run 4: Cell Editor Upgrade

Entry Criteria

Tasks

Deliverables

Verification Gate

Exit Criteria

Run 5: Entity-data-grid Evolution

Entry Criteria

Sub-run 5a: Row Auto-Publish Lifecycle

Sub-run 5b: Actions Column

Sub-run 5c: Search/Filter UI

Sub-run 5d: Pagination Modes

Sub-run 5e: Toolbar, Auto-Height, Drag-to-Scroll

Deliverables (Run 5 overall)

Exit Criteria

Run 6: Item-grid Integration

Entry Criteria

Tasks

Deliverables

Verification Gate

Exit Criteria

Run 7: Package API

Entry Criteria

Tasks

7a. Barrel Exports

7b. Dependencies

7c. Cleanup

Deliverables

Verification Gate

Exit Criteria

Run 8: Acceptance

Entry Criteria

Tasks

8a. Composition Stories

8b. VRT Baselines

8c. Documentation

Deliverables

Verification Gate

Exit Criteria

Run 9: canary-refactor Adaptation (Follow-on)

Entry Criteria

Tasks

Deliverables

Verification Gate