Slate v2 PortableText Harvest Plan

Objective: Produce a Slate v2 editor-harvest execution plan from local ../portabletext. Completion means the Portable Text harvest artifacts exist, license and output mode are recorded, all 502 inventory rows are classified, all 16 behavior rows are routed for slate-v2 as in-lane, split, or out-of-lane, current .tmp/slate-v2 owner coverage is mapped, downstream slate-plan rules are applied, and this file gives an accepted execution handoff without changing .tmp/slate-v2 code.

Goal plan: docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md

Template: docs/plans/templates/editor-harvest-plan.md

Primary template: docs/plans/templates/editor-harvest-plan.md

Applied packs:

• editor-harvest-plan
• editor-test-harvester for missing source harvest
• slate-plan downstream lane gates

Completion threshold:

• Score is 0.94, with every dimension at or above 0.91.
• Harvest report, inventory, and test-index exist under docs/editor-test-harvester/portabletext/.
• License mode is permissive and output mode is durable.
• All 502 inventory rows are classified in inventory.md; no uncertain row remains.
• All 16 behavior rows are accounted for in this plan as in-lane, split, or out-of-lane.
• Every in-lane or split raw-Slate row has a lane reason, current coverage mapping, target, proof route, and verification command.
• Downstream slate-plan gates are applied as planning-only boundaries: .tmp/slate-v2 is the owner, Plate/product rows stay out, current source wins, and no implementation/test edits happen in this pass.
• Accepted-plan execution handoff is present.
• Behavior rows use fresh invariant wording plus source-path provenance only.
• node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md passes.

Verification surface:

• Harvest artifacts: docs/editor-test-harvester/portabletext/report.md, inventory.md, test-index.md.
• Inventory command: rg --files ../portabletext piped through the editor-test-harvester include/exclude regex; exact count is 502.
• Owner coverage searches: rg over .tmp/slate-v2/packages/slate/test, slate-react/test, slate-browser/test, and .tmp/slate-v2/playwright for selection, fragment, insert, delete, paste, clipboard, history, collab, composition, IME, drag/drop, void, root, operation, and transform owners.
• Downstream lane skill: .agents/skills/slate-plan/SKILL.md.
• Plan integrity grep: rg for Harvest grounding, Lane contract, Full harvest row accounting, In-lane candidate matrix, Execution queue, Downstream lane application, and Accepted-plan execution handoff in this file.
• Final mechanical gate: node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md.

Constraints:

• This is planning/routing work only. Do not patch implementation code, tests, examples, package files, or build config.
• Preserve harvest license mode. Versioned plan output uses fresh invariant language plus source path and line provenance.
• Do not count Plate/product rows as Slate substrate rows. Split mixed rows.
• Browser, clipboard, selection, mobile, and IME rows need honest proof routes.

Boundaries:

• Harvest report: docs/editor-test-harvester/portabletext/report.md.
• Lane: slate-v2.
• Downstream skill: .agents/skills/slate-plan/SKILL.md.
• Owner boundary: .tmp/slate-v2 for raw Slate and slate-react behavior tests; Plate packages/docs for schema, list, annotation, serializer, plugin, toolbar, and Sanity bridge rows.
• Allowed edit scope in this pass: docs/editor-test-harvester/portabletext/** and docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md.
• Non-goals: implementation, package/runtime edits, GitHub comments, commits, pushes, PRs, and browser proof execution unless explicitly requested later.

Blocked condition: The only blocker would be losing access to ../portabletext, losing access to .tmp/slate-v2, or the user choosing a different lane owner for schema/list/plugin rows. None of those blockers is active.

Lane state:

• harvest_report: done at docs/editor-test-harvester/portabletext/report.md
• lane: slate-v2
• current_pass: closure review
• current_pass_status: complete
• current_pass_skill: .agents/skills/editor-harvest-plan/SKILL.md
• downstream_skill: .agents/skills/slate-plan/SKILL.md
• next_pass: accepted execution handoff
• goal_status: active until final mechanical gate and update_goal completion

Current verdict:

• verdict: accept plan
• score: 0.94
• next owner: slate-patch or slate-plan execution pass in .tmp/slate-v2
• reason: Portable Text gives strong raw pressure for roots, projected selection, object boundaries, insert/delete, drag/drop, IME, and remote rebasing; schema/list/plugin/serializer rows belong to Plate.

Completion rule:

• Do not call update_goal(status: complete) until all checklist items below are checked, verification evidence records the final commands, and check-complete passes.
• Do not create hook state for this goal. This file plus the active goal are the durable state.

Start Gates:

Gate	Applies	Evidence
Skill analysis before edits	yes	editor-harvest-plan, editor-test-harvester, and slate-plan skill bodies were read before writing the harvest and plan artifacts.
Active goal checked or created	yes	Active goal 019e6fd0-9a50-7120-ac35-068d11f61bcc exists for this exact Portable Text harvest plan.
Source of truth read before edits	yes	../portabletext/LICENSE, package.json, editor tests, selector tests, DOM structure tests, collaboration tests, and .tmp/slate-v2 owner test files were read before routing rows.

Work Checklist:

• Objective includes lane outcome, completion threshold, verification surface, constraints, boundaries, and blocked condition.
• Harvest report path, status, score, license mode, output mode, inventory counts, matrix rows, skips, next slice, and pass-state ledger recorded.
• Inventory and test-index existence/missing reasons recorded.
• Lane aliases normalized and lane registry row selected.
• Every harvest row counted as in-lane, out-of-lane, split, duplicate, skip, or unresolved.
• Every split row has lane-owned and out-of-lane portions separated.
• Current owner coverage searched in the target workspace before claiming covered or missing.
• Every in-lane row has lane reason, current coverage, action, target, proof route, and verification command or defer reason.
• Behavior-only rows use fresh-invariant; no copied source wording or mechanical translation entered versioned output.
• Downstream lane gates applied and recorded.
• Issue/claim accounting recorded, including explicit no-claim text when no claim changes.
• Accepted-plan execution handoff complete.

Completion Gates:

Gate	Applies	Required action	Evidence
Named verification threshold	yes	Verify harvest artifacts, row accounting, downstream gates, and final autogoal check.	Commands are listed in Verification evidence after execution.
Harvest artifacts current	yes	Verify report.md, inventory.md, and test-index.md exist and match the 502-row harvest.	docs/editor-test-harvester/portabletext/report.md records 502 rows, 321 runnable rows, and 211 indexed portable/portable-mixed rows.
Downstream lane gates applied	yes	Record slate-plan boundaries and owner rules.	Downstream lane application table records current-source, raw Slate owner, Plate split, and planning-only rules.
Issue or claim accounting changed	no	Record that no issue, PR, or claim ledger changed.	Issue and claim accounting section says no GitHub or ledger sync applies.
Accepted-plan handoff	yes	Emit execution queue and stop rule.	Accepted-plan execution handoff section is complete.
Goal plan complete	yes	Run node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md.	Final command is recorded in Verification evidence.

Phase / pass table:

Phase	Status	Evidence	Next
Harvest grounding	complete	MIT license read; harvest artifacts written.	done
Lane filter and row accounting	complete	16 behavior rows routed; 502 inventory rows classified.	done
Current owner coverage mapping	complete	.tmp/slate-v2 owner searches recorded.	done
Execution queue and proof routing	complete	Queue Q1 through Q6 written with targets and commands.	done
Downstream lane gate application	complete	slate-plan raw/Plate boundaries recorded.	done
Issue and claim accounting	complete	No issue or claim sync applies.	done
Accepted-plan execution handoff	complete	Handoff section names IDs, boundaries, commands, broad gate, and stop rule.	done
Closure review and final gates	complete	Mechanical checks recorded below.	final response

Confidence score:

Dimension	Weight	Score	Evidence	Cap hit
Harvest source readiness	0.15	0.97	MIT license, package scripts, 502-row inventory, 321 runnable rows, 211 portable/portable-mixed indexed rows.	none
Lane-filter completeness	0.25	0.94	16 behavior rows routed; 502 file rows classified; no uncertain row remains.	none
Current owner coverage mapping	0.25	0.91	.tmp/slate-v2 package and Playwright owner searches map every raw row to current owners or gaps.	no tests run because this is planning-only
Actionability of execution queue	0.20	0.93	Q1-Q6 list IDs, targets, proof kinds, focused commands, and stop rules.	browser rows still need execution when implementation starts
License/provenance discipline	0.15	0.95	Permissive license recorded; generated docs use source paths, line anchors, and fresh invariant wording.	none
Weighted total	1.00	0.94	Passes the 0.92 threshold and no dimension is below 0.85.	none

Harvest grounding:

Field	Value
Harvest report	docs/editor-test-harvester/portabletext/report.md
License mode	permissive, MIT from ../portabletext/LICENSE
Output mode	durable versioned harvest under docs/editor-test-harvester/portabletext/
Inventory status	502 rows classified in inventory.md; 321 runnable; no uncertain rows
Test-index status	211 runnable portable/portable-mixed rows indexed with 1979 line anchors
Matrix rows	16 behavior rows in report.md and routed below
Skips	95 Plate-owned rows, 182 harness rows, 14 skip rows; all accounted for
Next slice	Selection/fragment projection first, then object-boundary editing, then remote selection/history, then IME/focus/event authority

Lane contract:

Field	Value
Lane	slate-v2
Aliases	slate, raw-slate, Slate v2 substrate
Downstream skill	.agents/skills/slate-plan/SKILL.md
Owner boundary	.tmp/slate-v2/packages/slate, slate-react, slate-browser, and .tmp/slate-v2/playwright for raw behavior; Plate owns product APIs and package policy
Exclusions / split rules	Do not import Portable Text schema shape, behavior API names, list policy, annotation/decorator policy, serializers, plugins, toolbar, or Sanity bridge into raw Slate

Full harvest row accounting: All 502 file rows are classified in docs/editor-test-harvester/portabletext/inventory.md. The executable behavior routing rows are accounted for here.

Row	Source ref	Classification	Lane accounting	Reason
PT-SV2-01	boundary-equivalent selector utility tests	portable	in-lane	Logical point equivalence is raw selection substrate.
PT-SV2-02	fragment extraction across roots and containers	portable-mixed	split	Raw Slate owns fragment/root projection; Plate owns table/schema envelope policy.
PT-SV2-03	selected value slicing inside nested containers	portable-mixed	split	Raw Slate owns selection order/projection; Portable Text owns path serialization.
PT-SV2-04	overlap comparison with block and inline objects	portable	in-lane	Object endpoint comparison is raw Slate range logic.
PT-SV2-05	delete matrix	portable	in-lane	Delete across text, voids, objects, and blocks is raw transform/runtime behavior.
PT-SV2-06	insert block/object placement matrix	portable-mixed	split	Raw Slate owns placement and selection; Portable Text owns key generation and schema admissibility.
PT-SV2-07	copy/paste object and HTML matcher rows	portable-mixed	split	Raw Slate owns fragment/clipboard projection; Plate owns matchers and MIME policy.
PT-SV2-08	drag selection and self-drop rows	portable-mixed	split	Raw Slate owns native selection/drop validity; product drag handles belong above.
PT-SV2-09	DOM structure and drop path resolution	portable-mixed	split	Raw Slate owns path mapping/projection; Portable Text owns data attributes/renderers.
PT-SV2-10	selection after remote patches and dedup	portable	in-lane	Remote operation rebasing and selection emission are raw collab substrate.
PT-SV2-11	undo/redo collaboration rows	portable	in-lane	History over remote edits is raw Slate/history behavior.
PT-SV2-12	composition rows	portable-mixed	split	Raw Slate owns composition transport; Plate owns decorator/annotation inheritance policy.
PT-SV2-13	focus and keyboard event authority	portable-mixed	split	Raw Slate owns focus/selection/event authority; Portable Text behavior names stay out.
PT-SV2-14	applicable schema and sub-schema rows	portable-mixed	split	Raw content-root capability may matter; schema category policy is Plate/Portable Text.
PT-SV2-15	structured lists and textspec list rows	plate-owned	out-of-lane	List UX belongs to Plate.
PT-SV2-16	markdown/html/to-html/toolbar/plugin packages	plate-owned	out-of-lane	Serializers, renderers, toolbar, plugin, and Sanity bridge policy are not raw Slate.

In-lane candidate matrix:

Row	Source ref	Tag	Behavior invariant	Lane reason	Current coverage	Action	Target	Proof
PT-SV2-01	boundary-equivalent.test.ts line anchors	selection-boundary	Adjacent text-node edge points compare as one logical caret while preserving direction.	Raw point/range/projection law.	operations-contract.ts, selection-rebase-contract.ts, projections-and-selection-contract.test.tsx.	refactor-existing plus add named edge matrix if absent.	.tmp/slate-v2/packages/slate/test/selection-rebase-contract.ts and slate-react/projections tests.	bun test focused package files; Playwright only if DOM projection changes.
PT-SV2-02	selector.get-fragment.test.ts line anchors	fragment-roots	Fragment extraction unwraps only when selection is wholly inside one content root.	Multi-root/content-root projection.	multi-root-document.test.ts, projected-clipboard-contract.test.ts, content-root navigation tests.	create/refactor content-root fragment contract.	slate-react projected clipboard/selection tests and multi-root browser example.	focused package tests plus Playwright multi-root example grep.
PT-SV2-03	selector.get-selected-value.test.ts line anchors	nested-containers	Selection slicing across nested containers keeps logical order and trims only endpoints.	Root ordering and projected selection.	projections-and-selection, root interaction, projected clipboard.	create browser proof if package owners lack nested root range coverage.	slate-react projection tests and browser examples.	package tests plus browser drag/selection route.
PT-SV2-04	selector.is-overlapping-selection.test.ts line anchors	overlap-compare	Text, inline objects, and block objects share consistent inclusive endpoint overlap semantics.	Raw range compare.	view-selection-contract and root interaction resolver.	add focused unit if not already explicit.	slate-react/view-selection-contract.test.ts or slate package point/range tests.	bun test focused owner.
PT-SV2-05	event.delete.matrix.test.ts line anchors	delete-matrix	Delete handles text, words, blocks, void-at-edge, void-to-void, and inline object ranges without ghost nodes or invalid selection.	Raw transforms and runtime selection.	transforms/delete fixtures, operations-contract, generated browser stress.	refactor-existing names and fill object-boundary gaps.	slate/test/transforms/delete, operations-contract.ts, browser stress only for DOM-specific rows.	bun test slate transform files; browser stress grep for object deletion if touched.
PT-SV2-06	event.insert.blocks/block/inline-object tests	insert-placement	Block/object insertion at text edges, middle, empty block, and block object yields stable placement and final selection.	Raw insert transform and selection.	transforms insertNodes/insertText fixtures and transforms-contract.ts.	create compact placement/select contract if scattered.	slate/test/transforms-contract.ts and transform fixtures.	bun test slate transforms.
PT-SV2-07	event.paste.test.ts line anchors	clipboard-object	Clipboard fragments over objects and expanded selections preserve shape and caret.	Raw clipboard/fragment projection.	projected-clipboard-contract.test.ts and paste-html example.	split raw projection from Plate matcher policy.	slate-react projected clipboard; Playwright paste-html for DOM proof.	bun test projected clipboard; Playwright paste-html grep when touched.
PT-SV2-08	drag-selection and drag.drop tests	drag-drop	Dragging inside current selection uses that range; dragging outside selects the atom; self-drop into origin is suppressed.	Native selection/drop validity over roots and voids.	root-interaction-resolver and browser selection tests.	create Playwright rows for roots/voids if missing.	slate-react/root-interaction-resolver.test.ts and browser examples.	focused root interaction tests plus Playwright route grep.
PT-SV2-09	dom-structure and resolve-element-drop-position tests	dom-projection	DOM nodes and drop targets map by full projected path, including same-key nested roots.	Projection graph and content-root path law.	root-location-contract, content-root navigation, slate-void shell, projections tests.	add same-key nested root negative control.	slate-react projection/content-root tests.	bun test focused slate-react owners.
PT-SV2-10	selection-after-remote-patches and selection emit tests	remote-selection	Remote operations rebase local selection once and do not emit duplicate selection moves.	Raw collab/selection substrate.	collab-selection-stress-contract.ts, collab-document-state-contract.ts, projected-collab-substrate-contract.test.ts.	create/refactor remote selection matrix.	slate package collab tests and slate-react projected collab test.	bun test focused collab tests.
PT-SV2-11	undo-redo-collaboration tests	history-collab	Local undo removes only local edits after unrelated remote edits.	Raw history/collab behavior.	collab-history-runtime-contract.ts and slate-history tests.	refactor-existing or create explicit row.	slate/test/collab-history-runtime-contract.ts and slate-history tests.	bun test focused history/collab owners.
PT-SV2-12	composition.test.ts line anchors	ime-composition	Composition commit, cancel, replacement, and boundary crossing keep DOM selection and model text coherent.	Browser input transport.	slate-browser IME tests, composition-state-contract, generated stress rows.	mostly covered; add only missing boundary transport rows.	slate-browser/playwright-ime and slate-react composition-state tests.	bun test slate-browser core plus focused Playwright IME grep.
PT-SV2-13	focus and keyboard event tests	focus-keyboard	Focus and keyboard routing keep selection authority clear without losing native event semantics.	Raw focus/event substrate.	selection-controller, runtime provider, browser selection tests.	split raw event authority from Portable Text behavior API.	slate-react selection controller/runtime tests.	bun test focused slate-react owners.
PT-SV2-14	applicable-schema and sub-schema tests	content-root-capability	Content roots may expose allowed insertion capabilities without importing schema policy.	Only if Slate exposes root capability hooks.	content-root-navigation tests cover navigation, not schema.	defer raw API unless capability hook is accepted; route product policy to Plate.	no raw code target until API decision.	slate-plan API decision before tests.

Split rows:

Row	Source ref	Lane-owned part	Out-of-lane part	Owner / handoff
PT-SV2-02	fragment roots	Root projection and fragment boundaries.	Table/callout schema envelope serialization.	Raw Slate plus Plate serializer/schema.
PT-SV2-03	selected value nested containers	Projected selection order and endpoint trimming.	Portable Text path format and markDef pruning.	Raw Slate plus Plate/PT serializer.
PT-SV2-06	insert placement	Placement and selection.	Duplicate key generation and schema admissibility.	Raw Slate plus Plate/schema.
PT-SV2-07	paste objects	Fragment/clipboard projection.	Object matchers and Portable Text MIME.	Raw Slate plus Plate paste pipeline.
PT-SV2-08	drag/drop	Native selection/drop validity.	Drag handle UI and DnD product behavior.	Raw Slate plus Plate UI package.
PT-SV2-09	DOM structure	Full projected path mapping and void root shell.	Portable Text data attributes/renderers.	Raw Slate plus Plate renderers.
PT-SV2-12	composition	Browser composition transport.	Decorator/annotation inheritance policy.	Raw Slate plus Plate marks.
PT-SV2-13	focus/keyboard	Focus selection and event authority.	Portable Text behavior API naming.	Raw Slate plus Plate behavior API research.
PT-SV2-14	schema selection	Optional root capability surface.	Schema/category policy.	Slate API decision plus Plate schema owner.

Excluded or out-of-lane rows:

Row	Source ref	Reason	Owner
PT-SV2-15	structured lists and textspec list rows	List indentation/rendering is product/editor policy.	Plate list package.
PT-SV2-16	markdown/html/to-html/toolbar/plugin packages	Serializers, toolbar, plugins, Sanity bridge, and renderer packages are not raw Slate substrate.	Plate packages/docs.
Inventory harness bucket	182 harness rows	Fixtures/support/setup are accounted for but not behavior rows.	No execution owner.
Inventory skip bucket	14 skip rows	Non-editor/package rows do not pressure Slate v2.	No execution owner.

Coverage dedupe:

Candidate	Existing coverage	Decision	Evidence
IME base composition	slate-browser IME tests, composition-state-contract, generated stress rows	refactor-existing, not copy-now	Current owner search found direct IME/composition owners.
Delete and insert basics	slate transform fixtures plus transforms-contract.ts	refactor-existing before creating files	Current owner search found broad insert/delete fixture directories.
Multi-root navigation	multi-root-document.test.ts and content-root-navigation-contract.test.ts	create only nested/fragment gaps	Browser and package owners already exist.
Clipboard HTML paste	projected-clipboard-contract and paste-html browser example	split raw fragment from Plate matcher policy	Current owner search found both package and browser surfaces.
Collab history	collab-history-runtime-contract.ts	add remote-selection rows only if exact rebase/dedup gaps exist	Existing collab substrate found.

Execution queue:

ID	Action	Target	Proof kind	Focused verification	Notes
Q1	Apply PT-SV2-01 through PT-SV2-04.	.tmp/slate-v2/packages/slate/test/selection-rebase-contract.ts; .tmp/slate-v2/packages/slate-react/test/projections-and-selection-contract.test.tsx; view-selection-contract.test.ts; multi-root browser example.	unit plus browser where DOM projection changes.	cd .tmp/slate-v2 && bun test ./packages/slate/test/selection-rebase-contract.ts ./packages/slate-react/test/projections-and-selection-contract.test.tsx ./packages/slate-react/test/view-selection-contract.test.ts; focused Playwright multi-root grep if browser row changes.	First slice; highest value after recent selection/root bugs.
Q2	Apply PT-SV2-05 through PT-SV2-07.	slate transform delete/insert fixtures, transforms-contract.ts, projected-clipboard-contract.test.ts, paste-html example.	unit plus clipboard/browser proof.	cd .tmp/slate-v2 && bun test ./packages/slate/test/transforms-contract.ts ./packages/slate-react/test/projected-clipboard-contract.test.ts; Playwright paste-html grep when touched.	Keep matchers/MIME out of raw Slate.
Q3	Apply PT-SV2-08 and PT-SV2-09.	root-interaction-resolver.test.ts, content-root-navigation-contract.test.ts, slate-void-shell-contract.test.tsx, browser examples.	unit plus Playwright drag/selection proof.	cd .tmp/slate-v2 && bun test ./packages/slate-react/test/root-interaction-resolver.test.ts ./packages/slate-react/test/content-root-navigation-contract.test.ts ./packages/slate-react/test/slate-void-shell-contract.test.tsx; focused browser route grep.	Owns drag over roots/voids and full-path mapping.
Q4	Apply PT-SV2-10 and PT-SV2-11.	collab-selection-stress-contract.ts, collab-document-state-contract.ts, projected-collab-substrate-contract.test.ts, collab-history-runtime-contract.ts.	package unit.	cd .tmp/slate-v2 && bun test ./packages/slate/test/collab-selection-stress-contract.ts ./packages/slate/test/collab-history-runtime-contract.ts ./packages/slate-react/test/projected-collab-substrate-contract.test.ts.	No provider-specific Yjs policy in raw Slate.
Q5	Apply PT-SV2-12 and PT-SV2-13 only for exact gaps.	slate-browser IME tests, composition-state-contract, selection controller/runtime tests.	browser input plus unit.	cd .tmp/slate-v2 && bun test ./packages/slate-browser/test/core/playwright-ime.test.ts ./packages/slate-react/test/composition-state-contract.test.ts ./packages/slate-react/test/selection-controller-contract.test.ts.	Avoid importing Portable Text behavior API names.
Q6	Route PT-SV2-14 through PT-SV2-16 to Plate backlog.	Plate schema/list/serializer/plugin packages and docs.	package/docs when Plate work starts.	No .tmp/slate-v2 command.	Keep raw Slate unopinionated.

Issue and claim accounting:

• Fixed issues: none; this pass is planning-only.
• Improved issues: none; no GitHub issue or ledger claim changed.
• Related issues: none linked in the prompt.
• PR reference: none.
• Claim sync rule: If a later execution pass claims coverage for public Slate issues, sync the relevant issue/claim ledger in that pass. This plan itself changes no issue status.

Downstream lane application:

Gate	Status	Evidence
downstream skill read	complete	.agents/skills/slate-plan/SKILL.md was read before lane routing.
lane-specific completion gates applied	complete	Raw Slate owns substrate behavior; Plate owns product/schema/plugin policy; current .tmp/slate-v2 tests are mapped before new targets.
implementation boundaries recorded	complete	This pass edits only harvest/plan docs. Execution queue targets .tmp/slate-v2 for a later accepted pass.
verification commands recorded	complete	Each queue row names focused bun/Playwright commands; broad final gate is cd .tmp/slate-v2 && bun check.

Accepted-plan execution handoff:

• read-first plan path: docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md
• requested lane: slate-v2 raw substrate tests from Portable Text harvest.
• exact execution queue IDs: Q1 first, then Q2, Q3, Q4, Q5 only for exact gaps, Q6 for Plate backlog routing.
• implementation boundaries: edit .tmp/slate-v2 only in a later execution pass; do not edit .tmp/slate-v2 in this planning pass; do not pull Portable Text schema/list/plugin policy into raw Slate.
• focused verification commands: the commands in Execution queue, scoped to changed owners.
• broad final gate: cd .tmp/slate-v2 && bun check; use bun check:full only when browser architecture or release-quality browser claims are changed.
• issue/claim sync rule: sync claim ledgers only if the execution pass changes a public issue/claim status.
• stop rule: stop and reroute if a row requires schema/list/plugin/product semantics rather than raw editor substrate behavior.

Findings:

• Portable Text has 502 test/support inventory rows; 321 contain test/scenario calls.
• The raw Slate value is concentrated in 180 portable rows and 31 portable-mixed rows.
• The highest-value Slate rows are selection boundary equivalence, nested content-root fragment selection, delete/insert object boundaries, drag/drop over selected ranges, DOM full-path projection, and remote selection rebasing.
• The most tempting wrong move is importing Portable Text schema/behavior API shape into raw Slate. Do not do that.

Decisions and tradeoffs:

• Use Portable Text as a behavior oracle, not as an API model.
• Prioritize Q1 because recent Slate v2 bugs were selection/root/projection failures.
• Keep Plate-owned rows explicit instead of pretending raw Slate should own lists, schema categories, decorators, annotations, serializers, toolbar, or Sanity bridge policy.
• Do not run .tmp/slate-v2 tests in this pass because no .tmp/slate-v2 code changed.

Error attempts:

Error / failed attempt	Count	Next different move	Resolution
Initial generated inventory narrowed to 359 rows because shell regex escaping was wrong inside the generator.	1	Generate from rg --files and apply the include/exclude regex in JS.	Regenerated artifacts with 502 rows, matching the exact inventory command.

Verification evidence:

• Harvest files generated: docs/editor-test-harvester/portabletext/report.md, inventory.md, test-index.md.
• Inventory count generated from JS-filtered rg --files: 502 rows, 321 runnable rows, 211 portable/portable-mixed indexed rows, 1979 line anchors.
• Artifact presence check passed: harvest-files-ok.
• Exact inventory command count passed: 502.
• Report grep passed for status: done, score: 0.93, license_mode: permissive, Total inventory rows: 502, Portable rows: 180, Pass-State Ledger, and Next Slice.
• Plan grep passed for Harvest grounding, Lane contract, Full harvest row accounting, In-lane candidate matrix, Execution queue, Downstream lane application, and Accepted-plan execution handoff.
• Placeholder grep found no template markers, unresolved status labels, active-progress labels, or placeholder-only lines in the plan/report.
• Mechanical completion passed: [autogoal] complete: docs/plans/2026-05-28-slate-v2-portabletext-harvest-plan.md.

Reboot status:

Question	Answer
Where am I?	Closure verification for the Portable Text to Slate v2 harvest plan.
Where am I going?	Run artifact checks and the autogoal mechanical completion check, then close the active goal.
What is the goal?	A complete planning-only Slate v2 test harvest plan from ../portabletext with all rows routed and no .tmp/slate-v2 code edits.
What have I learned?	Portable Text is excellent pressure for root/projection/object-boundary tests; it is a bad source for raw Slate API shape.
What have I done?	Generated the harvest report, full inventory, test index, and filled this lane plan.

Timeline:

• 2026-05-28T21:54:10.716Z Goal plan created.
• 2026-05-28T22:00:00Z Portable Text license, package scripts, and editor test source were read.
• 2026-05-28T22:05:00Z Harvest artifacts generated and corrected to 502 inventory rows.
• 2026-05-28T22:10:00Z Slate v2 lane plan filled with row accounting, coverage mapping, and execution handoff.

Open risks:

• No material risk for the planning artifact. Execution risk remains in later browser proof for drag/drop, clipboard, and multi-root selection rows.