slate v2 legacy evidence lane

Objective: Add the first real editor comparison lane to /Users/zbeyens/git/plate-2/benchmarks/editor for .tmp/slate-v2 versus ../slate only. Completion means Evidence Kit benchmark results contain normalized Slate v2 and legacy Slate rows from the Slate v2 legacy-compare artifact, other editor runtimes stay deferred, docs/search expose the lane, root/editor verification passes, and this autogoal plan passes its closure check.

Goal plan: docs/plans/2026-05-27-slate-v2-legacy-evidence-lane.md

Template: docs/plans/templates/major-task.md

Primary template: docs/plans/templates/major-task.md

Applied packs:

• autogoal: durable completion gates and final closure check.
• major-task: benchmark comparison, tradeoffs, and evidence record.
• task: direct repo implementation and verification.

Major source:

• type: local benchmark artifact plus Evidence Kit package source
• id / link: .tmp/slate-v2/tmp/slate-react-huge-document-legacy-compare-benchmark-compare-v2DefaultRenderAuto-v2DomPresent-blocks-5000-iters-5-ops-10-split-selection-no-profile.json
• title: Slate React huge document legacy compare
• decision to make: first comparison lane is Slate v2 versus legacy Slate only
• decision criteria: exact .tmp/slate-v2 and ../slate repos, structured Evidence Kit rows, visible mixed results, no ProseMirror/Lexical/Tiptap runtime claims
• likely files / surfaces: benchmarks/editor/src/index.mjs, benchmarks/editor/benchmarks/slate-v2-legacy-benchmark.mjs, benchmarks/editor/benchmarks/results/slate-v2-legacy-latest.json, benchmarks/editor/docs/perf/*, benchmarks/editor/iterations/*
• browser surface: static Evidence Kit docs at benchmarks/editor/docs/perf/index.html
• highest-leverage owner: editor benchmark evidence lab

Major lane:

• lane: editor benchmark evidence
• output type: implementation plus generated benchmark/docs artifacts
• implementation expected: yes
• affected packages / surfaces: benchmarks/editor standalone npm package and root bench:editor:* command surface through existing scripts
• dominant risk: treating one huge-document artifact as a blanket editor verdict

Completion threshold:

• Fresh local Slate benchmark run records currentRepo as /Users/zbeyens/git/plate-2/.tmp/slate-v2 and legacyRepo as /Users/zbeyens/git/slate.
• Evidence Kit writes benchmarks/editor/benchmarks/results/slate-v2-legacy-latest.json with normalized rows for both slate-v2:* and legacy-slate:*.
• The result contains 52 rows: 13 common millisecond metrics across legacy-slate:chunk-off, legacy-slate:chunk-on, slate-v2:default-render-auto, and slate-v2:dom-present.
• ProseMirror, Lexical, Plate, and Tiptap runtime adapters remain deferred; no placeholder runtime rows are emitted for them.
• Root/editor verification passes, perf docs/search expose the lane, and node .agents/rules/autogoal/scripts/check-complete.mjs docs/plans/2026-05-27-slate-v2-legacy-evidence-lane.md passes.

Verification surface:

• Fresh benchmark command in /Users/zbeyens/git/plate-2/.tmp/slate-v2: REACT_HUGE_COMPARE_LEGACY_REPO=../../../slate REACT_HUGE_COMPARE_ITERATIONS=5 REACT_HUGE_COMPARE_TYPE_OPS=10 REACT_HUGE_COMPARE_SPLIT_SELECTION=1 REACT_HUGE_COMPARE_DISPOSE_DELAY_MS=0 REACT_HUGE_COMPARE_SURFACES=v2DefaultRenderAuto,v2DomPresent bun run bench:react:huge-document:legacy-compare:local
• Evidence Kit command in /Users/zbeyens/git/plate-2: npm run bench:editor:check
• Search command in /Users/zbeyens/git/plate-2: npm run bench:editor:search -- slate-v2 legacy
• Formatting command in /Users/zbeyens/git/plate-2: pnpm exec biome check package.json benchmarks/editor docs/plans/2026-05-27-slate-v2-legacy-evidence-lane.md --fix

Constraints:

• Start from local repo evidence before external claims.
• Keep the first comparison narrow: .tmp/slate-v2 versus ../slate.
• Preserve unsupported, slow, and uncomfortable rows instead of reducing the result to a winner.
• Do not restore the deleted benchmark app/template lab.
• Do not commit or open a PR.

Boundaries:

• Source of truth: the refreshed Slate v2 legacy-compare JSON artifact and the Evidence Kit benchmark-result schema.
• Allowed edit scope: benchmarks/editor/**, root/package benchmark script wiring already present, and this goal plan.
• External sources: not used; local benchmark artifact and local Evidence Kit package code settle the implementation.
• Browser surface: static perf docs were regenerated; Browser plugin reload was unavailable in this turn, so proof is file/search based.
• Tracker sync: not applicable; no external issue or PR requested.
• Non-goals: ProseMirror, Lexical, Plate, Tiptap adapters; cross-editor browser interaction fixtures; broad editor verdicts.

Blocked condition:

• Autonomous work would stop only if the Slate v2 legacy-compare script could not produce an artifact for .tmp/slate-v2 versus ../slate, or if Evidence Kit rejected the normalized benchmark rows. Neither blocker occurred.

Major state:

• task_type: major
• task_complexity: major
• current_phase: closeout
• current_phase_status: complete
• next_phase: final response
• goal_status: ready-for-closure

Current verdict:

• verdict: lane implemented
• confidence: high for Evidence Kit integration, medium for performance interpretation
• next owner: benchmark maintainer
• reason: the lane is exact and searchable, but one 5,000-block workload is not a global editor-performance answer

Completion rule:

• Every checklist item is closed.
• Every gate has evidence.
• This file plus the active autogoal are the durable state.

Start Gates:

Gate	Applies	Evidence
major-task loaded	yes	Skill instructions read before implementation.
Active goal checked or created	yes	Active goal created for Slate v2 legacy Evidence Kit lane.
Source of truth read before analysis	yes	Read Evidence Kit source and Slate v2 legacy-compare artifact/script.
Major lane selected	yes	Lane selected: editor benchmark evidence.
Decision criteria stated	yes	Criteria listed in Major source and Completion threshold.
Existing repo patterns / prior decisions checked	yes	Reused existing Evidence Kit rows, fuzzer, docs, scope, and package scripts.
Helper stack selected	yes	Used local Node scripts plus Evidence Kit CLI; no new framework.
External research decision recorded	yes	External research not used because local artifacts settle this lane.
Implementation expectation recorded	yes	Implementation expected and completed in benchmarks/editor.
Workspace authority selected	yes	Authority is /Users/zbeyens/git/plate-2 plus .tmp/slate-v2 benchmark artifact source.
Branch / PR expectation decided	yes	No commit or PR requested.

Work Checklist:

• Objective includes outcome, completion threshold, verification surface, constraints, boundaries, and blocked condition.
• Major source records source type, id/link, title, decision type, expected outcome, decision criteria, likely files/packages/surfaces, browser surface, and highest-leverage owner.
• Current state is mapped before proposing a new architecture, migration, benchmark, or plan.
• Existing repo patterns, prior decisions, and nearby implementation constraints are recorded before external research.
• External docs or source are used only where repo evidence does not settle the question, or not-applicable reason is recorded.
• Options, recommendation, tradeoffs, blast radius, and rejection reasons are recorded.
• Facts, inference, and recommendation are separated.
• Review or pressure lenses are selected and completed, or marked not-applicable with reason.
• If implementation happens, touched-surface packs cover docs, browser, package/API, or agent-native surfaces as needed.
• Workspace authority recorded: every proof command names the cwd/tool that owns the analyzed or changed behavior.
• Accepted/actionable review findings are fixed or explicitly rejected with evidence.

Completion Gates:

Gate	Applies	Required action	Evidence
Named verification threshold	yes	Run benchmark, Evidence Kit check, search, and formatting	Fresh Slate benchmark passed; npm run bench:editor:check passed; search returns Slate lane rows; Biome passed.
Current-state source audit	yes	Map current owner, boundaries, constraints, and affected surfaces	Current owner is benchmarks/editor; artifact source is .tmp/slate-v2; legacy source is ../slate.
Decision criteria closure	yes	Mark each criterion satisfied, narrowed, rejected, or blocked with evidence	Satisfied for Slate v2 vs legacy; narrowed other editors to deferred.
Options / tradeoffs / rejection record	yes	Record viable options, chosen recommendation, and why alternatives lose	Chose artifact normalizer over new app; rejected cross-editor adapters for this first lane.
Review / pressure pass	yes	Run selected reviewer/lens or record reason	Manual pressure pass: result is mixed and cannot be sold as a blanket Slate v2 win.
Review findings closure	yes	Fix or explicitly reject accepted/actionable findings and record closure proof	Exact artifact rerun fixed stale embedded currentRepo concern.
External-source audit	no	Cite official/local clone/external sources when used, or record reason	No external sources used; local artifact and package source are authoritative.
Implementation gates	yes	If code changed, close primary-template and touched-surface gates	Parser, script, fuzzer, corpus, docs, generated results, and iteration note completed.
Final handoff contract	yes	Record recommendation, evidence, caveats, residual risk, and next owner	See Final handoff contract section.
Final lint	yes	Run scoped equivalent when files changed	pnpm exec biome check package.json benchmarks/editor docs/plans/2026-05-27-slate-v2-legacy-evidence-lane.md --fix passed.
Goal plan complete	yes	Run autogoal closure script	Closure script runs after this plan update.

Phase / pass table:

Phase	Status	Evidence	Next
Intake and source read	complete	Read plan, Evidence Kit source shape, Slate artifact/script, package scripts	Current-state map
Current-state map	complete	Existing lab had readiness rows only; no real editor runtime rows	Options
Options and recommendation	complete	Chose source-artifact normalizer and dedicated result file	Review
Review / pressure pass	complete	Mixed-result caveat recorded; no blanket win claim	Implementation
Implementation or plan artifact	complete	Added parser, benchmark writer, fuzzer/corpus coverage, docs note	Verification
Verification	complete	Benchmark, root check, search, and Biome passed	Closeout
Closeout	complete	Plan updated for autogoal closure	Final response

Findings:

• Evidence Kit indexes every benchmarks/results/*latest.json, so a dedicated slate-v2-legacy-latest.json is enough for docs/search without changing the core benchmark result.
• The refreshed artifact is exact to the requested repos: currentRepo=/Users/zbeyens/git/plate-2/.tmp/slate-v2 and legacyRepo=/Users/zbeyens/git/slate.
• Rows are mixed: readyMs strongly favors Slate v2, while selectAllMs strongly favors legacy chunking.

Decisions and tradeoffs:

• Chosen: normalize the existing Slate v2 browser benchmark artifact into Evidence Kit rows.
• Rejected: recreate the old benchmark app lab. That adds UI surface without improving the first evidence claim.
• Rejected for this pass: ProseMirror, Lexical, Plate, and Tiptap adapters. They need their own source-backed fixture work.
• Recommendation: treat this as a reliable lane for one huge-document workload, not as a broad editor shootout.

Implementation notes:

• benchmarks/editor/src/index.mjs exports Slate legacy compare artifact helpers and normalizes common millisecond metrics into Evidence Kit rows.
• benchmarks/editor/benchmarks/slate-v2-legacy-benchmark.mjs writes and checks benchmarks/results/slate-v2-legacy-latest.json.
• benchmarks/editor/package.json wires the new benchmark into bench:evidence and check.
• benchmarks/editor/test/fuzz/core-fuzz.mjs and benchmarks/editor/test/fixtures/corpus.json cover row normalization.
• benchmarks/editor/README.md and benchmarks/editor/iterations/001-slate-v2-legacy-evidence.md document the lane and its caveat.

Review fixes:

• Fixed the artifact exactness issue by rerunning the Slate v2 benchmark from .tmp/slate-v2 with REACT_HUGE_COMPARE_LEGACY_REPO=../../../slate.
• Kept V2-only metrics out of the first comparison so each fixture has both Slate v2 and legacy rows.
• Kept other editor runtimes deferred; no fake unsupported rows were added for them.

Error attempts:

Error / failed attempt	Count	Next different move	Resolution
Browser plugin reload unavailable after tool discovery	1	Use regenerated docs/search artifacts as proof	Static docs and search proof passed; no browser-only behavior changed.

Verification evidence:

• In /Users/zbeyens/git/plate-2/.tmp/slate-v2, fresh benchmark command passed and wrote the 5,000-block compare artifact with exact repo paths.
• In /Users/zbeyens/git/plate-2/benchmarks/editor, npm run bench:evidence passed and wrote 52 Slate compare rows.
• In /Users/zbeyens/git/plate-2, npm run bench:editor:check passed. npm printed config warnings for pnpm-oriented settings; warnings only.
• In /Users/zbeyens/git/plate-2, npm run bench:editor:search -- slate-v2 legacy returned slate-react-huge-document-legacy-compare benchmark fixtures from benchmarks/results/slate-v2-legacy-latest.json.
• In /Users/zbeyens/git/plate-2, scoped Biome check with --fix passed.

Final handoff contract:

• Recommendation: use this as the first Evidence Kit comparison lane and defer other editor runtimes until each has its own source-backed fixture.
• Confidence: high for integration; medium for benchmark interpretation.
• Evidence: 52 rows, 13 fixtures, 4 surfaces, exact .tmp/slate-v2 and ../slate repo paths.
• Tests / commands: fresh Slate benchmark, npm run bench:editor:check, npm run bench:editor:search -- slate-v2 legacy, scoped Biome.
• Browser proof: static docs regenerated; Browser plugin reload unavailable in this turn.
• PR / tracker: none requested.
• Caveats: local machine benchmark noise; one huge-document workload; mixed result where Slate v2 wins readiness/full-document rows and legacy chunking wins some selection rows.
• Next owner: benchmark maintainer.

Timeline:

• 2026-05-27T18:43:52.543Z Major-task goal plan created.
• 2026-05-27T18:52:00Z Fresh .tmp/slate-v2 versus ../slate benchmark artifact created.
• 2026-05-27T18:53:45Z Root editor Evidence Kit check passed.
• 2026-05-27T18:54:00Z Perf docs/search and formatting proof completed.

Reboot status:

Question	Answer
Where am I?	Closeout complete
Where am I going?	Final response after autogoal closure
What is the goal?	First real Evidence Kit editor comparison lane: .tmp/slate-v2 versus ../slate
What have I learned?	The lane is meaningful but mixed: Slate v2 wins readiness/full-document rows; legacy chunking wins some selection rows
What have I done?	Implemented parser, benchmark writer, tests, generated result/docs, and verification

Open risks:

• Local benchmark noise remains; repeat runs can move absolute numbers.
• This is a huge-document Slate-only lane, not a general editor ranking.
• Browser plugin reload was unavailable, so visualization proof is regenerated docs/search rather than an in-app browser screenshot.