[objectstack-fleet]

Requested by Warren · project thread

Part of #25467

Clause-②: no
Metadata-protocol: none

维护者速读

• 改了什么：把 protocol.ts 里“数据 CRUD”一节中四个单条写入方法（validateData、createData、updateData、deleteData）的方法体，经 E1 宿主机制原样搬进新模块 protocol-data-writes.ts；四个方法仍留在类上（签名与文档不变），方法体变成一行转发。
• 为什么改：[maintainer] refactor(metadata-protocol): split src/protocol.ts (33,065 lines, +10,169 in 30 days) along its ObjectStackProtocolImplementation seams, one move-only slice per PR #25467 裁决 E 的 E2 第 3 节，在席位声明的 protocol.ts 冻结日内。第 2 节（查询校验）已由 refactor(metadata-protocol): move the query validation out of protocol.ts into protocol-query-validation.ts (#25467 cut 3) #25538 搬完，无可切；第 3 节其余四个方法体按席位裁决 A（6594932448）暂留类中，原因见下文。
• 风险与回滚：低。只做搬移：工具逐个证明 4 个方法体与原方法体在“this. 读作 host.”之外 AST 与文本全同；对外 index.d.ts（ESM 与 CJS）与 main 构建出的逐字节相同。回滚即整体 revert。
• 先看：packages/metadata-protocol/src/protocol-data-writes.ts 文件头；protocol.ts 里 4 个方法的转发体；scripts/engine-double-contract.seams.json 两行路径。
• 两分钟验证：cd packages/metadata-protocol && node scripts/host-split.mjs check origin/main updateData protocol-data-writes.ts，预期 "ok":true,"textIdentical":true。
• 需要你做的：无，席位落地。

What changes

Step E2, section 3 "data CRUD" of ruling E (6537939895), on the protocol.ts freeze day seat 4 declared (6594835694 on #27737). It is a pure application of the E1 host pattern: no logic changes, and no new rewrite kind.

member	moved to	rewrite sites (all this.x to host.x)
validateData	protocol-data-writes.ts	2
createData	protocol-data-writes.ts	3
updateData	protocol-data-writes.ts	5
deleteData	protocol-data-writes.ts	3

Each method keeps its docblock, modifiers and signature on the class; its body becomes return fn(protocolHost(this), …). ProtocolHost and the class's static block gain 3 entries, all call-time closures: assertObjectRegistered, assertVersionMatch and probeRecord (engine was already on the host). node scripts/host-split.mjs check origin/main MEMBER protocol-data-writes.ts reads ok: true, textIdentical: true for all four, and the tool refused none. Negative control, with deleted === false changed to !== in deleteData's moved body, then restored from HEAD (blob hash equal to the disk hash afterwards):

PROBLEM protocol-data-writes.ts:286: ExclamationEqualsEqualsToken where the base has EqualsEqualsEqualsToken (`===`)
PROBLEM protocol-data-writes.ts: deleteData's body text differs from the base body rewritten and re-indented, first at body line 24

protocol.ts drops one import name only updateData's body read (byIdUpdateRecord; its re-export stays) and gains the import of the four functions beside the 1b imports.

Section 2 has nothing to cut

The order named section 2 (query validation) first. On main c9c68288 its 15 members (assertObjectRegistered through servesExpansionTarget, lines 4881 to 5613) are all stubs of the per-member series' fn.call(host, …) shape, with their bodies in protocol-query-validation.ts since #25538 (cut 3, merged 2026-10-23). No member in that range still carries a body. The seat re-ordered this thread to section 3.

Four bodies of section 3 stay in the class (seat ruling A, 6594932448)

This cut is under ruling E's 1,000-line guide on the seat's ruling. A trial move of all eight bodies measured two gates that read the section by path:

• findData, getData, cloneData, probeRecord stay. They hold all 6 of the query-options erasure sites scripts/query-options-erasure-baseline.json grandfathers for protocol.ts (findData 3, getData, cloneData, probeRecord 1 each). In a new module they fail pnpm lint and the ratchet, and listing the module is the move that baseline forbids (its key set only shrinks against the merge base). Cut 4 kept probeRecord in the class for the same reason. The seat is filing the card that types those erasures; these four move after it.
• updateData and deleteData move, and the engine-double-contract ledger follows them. Both are consumer seams the ledger keys by file. scripts/engine-double-contract.seams.json is rewritten by --write (two file values, nothing else), and the self-test's one wiring assertion names the module instead of protocol.ts. The rule judged is unchanged: before the ledger move the run read SEAMS_RETAINED four times and the self-test failed on the old path (exit 1); after it both exit 0, and the moved bodies still pass REFUSES (they throw recordNotFoundError).

Tests

No pin needed an edit. The read-verb pin, the lookup-artifact and hydrate caller-set pins and the system-context census read nothing in the four moved bodies. The query-transport-dialect pin and core's second-object-read-exposure pin read findData and getData, which stay.

The published .d.ts

Built from main 7ba8b11f5 and from this head with the same dependency build. index.d.ts is byte-identical: raw sha256 3a15d153…f48075 on both sides, ESM equal to CJS, so no normaliser is needed.

A measurement worth keeping: on the first base (c9c68288, before #29730), getMetaTypes()'s inferred return type made tsc's declaration emit print a docblock of protocol.ts chosen by character offset (the comment leak 1a met). Any net shift of the text above the class changed index.d.ts, and only a net-zero shift kept it identical (probed at 12 to 28 added characters; only 18, the exact compensation, matched). #29730 gave getMetaTypes() a named return type, and on the merged base the import stands in the import block with index.d.ts byte-identical. The leak no longer constrains later E2 cuts.

Serial

Merged origin/main twice: over #29730 (the getMetaTypes() return type) and over #29753 (seat 4's flow-read docblock lines), both landed during the freeze as the seat's post allowed. git merge was clean both times, and host-split check was re-run against the merged main for all four members. Seat 2's #29673 edits members outside this section and this diff does not touch them.

Verification

On ae528b8277:

• Package typecheck and test through os-verify-lock: VERDICT command-exit 0; Test Files 310 passed, 4 skipped (314); Tests 22301 passed, 26 skipped (22327).
• eslint packages/metadata-protocol scripts/check-engine-double-contract.mjs --no-inline-config exits 0; pnpm check:query-options-erasure holds (44 sites in 15 files, none new).
• node scripts/pm/dispatch-gates.mjs --commands derived 74 commands; --ran reports 74 derived, 73 run, 1 NOT-MEASURED, 0 UNRUN. Each run family exits 0 (check:pm-dispatch-gates 1945 fast-tier cases). The NOT-MEASURED one is check:dual-build-cjs-loads, which needs every package built (exit 3, PREREQUISITE NOT MET), so its reading is CI's.

Human Minimum

• 1. Schemas start as Zod; packages/spec holds no runtime logic. (untouched)
• 2. Every name I added uses the casing for its kind.
• 3. No formatter ran; pnpm lint passes (package and changed script locally; CI runs the whole).
• 4. One issue, one branch off an up-to-date main, one pull request.
• 5. A change that publishes carries a changeset, and both declaration lines are filled in here and in the changeset.
• 6. packages/spec touched: n/a.
• 7. No hand edits to content/docs/references/, packages/*/CHANGELOG.md or content/docs/releases/.
• 8. Title, body and comments are in English (the 维护者速读 block is the dispatch template's).

---

Generated by Claude Code

[github-actions]

📓 Docs Drift Check

This PR changes 1 package(s): @objectstack/metadata-protocol, touching 22 documentable anchor(s).

12 hand-written doc(s) NAME something this change touched and may need an implementation-accuracy re-verification:

• content/docs/api/client-sdk.mdx (via meta.publishItem (sdk, the route ledger binds it to POST /api/v1/meta/:type/:name/publish, selected by route anchor /:type/:name/publish), publishItem (sdk, the bare tail of client method meta.publishItem, bound to POST /api/v1/meta/:type/:name/publish))
• content/docs/api/metadata-api.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/concepts/metadata-lifecycle.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/data-modeling/currency.mdx (via /:object/import/jobs (route, bridged from symbol createData — its route source's handler names it))
• content/docs/data-modeling/import-mappings.mdx (via /:object/import/jobs (route, bridged from symbol createData — its route source's handler names it))
• content/docs/data-modeling/validation.mdx (via validateData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function))
• content/docs/kernel/contracts/data-engine.mdx (via createData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function), updateData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function))
• content/docs/kernel/contracts/metadata-service.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/kernel/services-checklist.mdx (via createData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function), deleteData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function), updateData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function))
• content/docs/permissions/system-context.mdx (via /:object/import/jobs (route, bridged from symbol createData — its route source's handler names it), /import/jobs/:jobId/undo (route, bridged from symbol deleteData — its route source's handler names it))
• content/docs/protocol/kernel/error-handling.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/protocol/objectql/state-machine.mdx (via /:object/import/jobs (route, bridged from symbol createData — its route source's handler names it), /import/jobs/:jobId/undo (route, bridged from symbol deleteData — its route source's handler names it))

⛔ 5 release-owned page(s) also name something this change touched. These are read-only:

• content/docs/releases/v17/17-0.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/releases/v17/17-1.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/releases/v17/17-2.mdx (via /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/releases/v17/17-4.mdx (via createData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function), /:type/:name/publish (route, bridged from symbol createData — its route source's handler names it; bridged from symbol deleteData — its route source's handler names it))
• content/docs/releases/v17/17-6.mdx (via validateData (symbol, a method of class ObjectStackProtocolImplementation; a top-level function))

content/docs/releases/ is RELEASE-OWNED (AGENTS.md "Documentation Guardrails"): release
notes are written centrally at release time, and a code PR that edits them is the exact PR
that guardrail exists to stop. They are still audited — read-only. If one of them is actually
wrong, file an issue or open a dedicated docs-only PR; do not edit it here.

What this run could not see

• 1 cross-cutting symbol(s) contributed no route anchor: updateData (4 routes)
• the SDK route bridge reached 61 of 218 client-bound route-ledger rows — the other 157 have no registrar path: tail to select them, so pages documenting THEIR client methods cannot appear above, on this or any run. Of those 157: 0 are remediable by widening that discovery convention (an in-repo file declares the path; the convention did not scan it); 55 are structural — on a ledger where NOT ONE row is declared in-repo, so no discovery change reaches them at any price; 102 are undecided (no in-repo declaration, on a ledger that has other in-repo registrars — absence and an unreadable spelling are not distinguishable here). The rows themselves: node scripts/docs-audit/affected-docs.mjs --bridge-coverage
• a page that states a rule by its inputs shares no identifier with the emitter that implements the rule, so an emitter-only diff cannot list it — not on this run and not on any run. Measured on fix(driver-sql): emit varchar(maxLength) for a text field a declared index keys on #11430: content/docs/protocol/objectql/types.mdx documents the text-family column mapping by the ObjectQL type names it maps FROM (text / textarea / html) while the diff changed createColumn; it went unlisted, and it was the page that diff falsified, in four places. No shared token exists to detect this on, so a rule your change carries has to be re-read by hand in the pages that restate it.
• a key NAME is not a key, so the hand re-read the line above prescribes can land on the wrong schema. The same spelling is authorable on one governed type and a [REMOVED] tombstone on another for each of active, aria, joins, objects, template, tools and version (censused on [finding] tools is a key on BOTH AgentSchema (tombstoned, dead) and SkillSchema (live, cloud-attested), so a name-based search attributes skill examples to the agent key — it produced a false stop-the-line alarm on PR #19059 #19093 over the liveness ledger's governed types, top-level keys); nothing in a search result distinguishes the two, so a grep hit on a LIVE example reads as evidence about the DEAD key. Measured on fix(spec): the agent.tools liveness row says dead — it claimed live on a key the schema tombstoned #19059: content/docs/ai/agents.mdx was reported as contradicting the agent.tools tombstone over its tools: example at :161, which is inside the defineSkill({ block opened at :155 — the page was already correct. Settle ownership by PARSING the value against both schemas, never by the name: that literal PASSES SkillSchema, and as an AgentSchema it FAILS at tools with the tombstone prescription. ⛔ These names are not the whole class — a key retired through a .strict() guidance map leaves no tombstone in the walked shape and none of them here (tool.category, live as AIToolDefinition.category).

Coarse fallback — 10 page(s) merely mention a changed package (the pre-#9192 predicate, kept for the deliberately-wide backstop): node scripts/docs-audit/affected-docs.mjs --json 601832b1a667992e73d54bf11643849836a6f778 → packageMentionDocs.

Which tree this was computed on

This run read content/docs from f586fdcb8beabe92b6b6c8cab1090af34df9984d — the merge of head ae528b82771c30d29c90b7370417750de2a48d61 into base 601832b1a667992e73d54bf11643849836a6f778, which is what actions/checkout gives a pull_request run. Not the PR head.

A worktree cut from an older main holds a different content/docs, so re-deriving there can legitimately return a different list — that is a different tree, not a wrong row. To answer on the same tree:

# while this PR is open — GitHub drops the merge commit once it closes
git fetch origin f586fdcb8beabe92b6b6c8cab1090af34df9984d && git checkout f586fdcb8beabe92b6b6c8cab1090af34df9984d
# afterwards, rebuild it from the two parents, which stay fetchable
git fetch origin 601832b1a667992e73d54bf11643849836a6f778 ae528b82771c30d29c90b7370417750de2a48d61 && git checkout -B drift-repro 601832b1a667992e73d54bf11643849836a6f778 && git merge --no-ff ae528b82771c30d29c90b7370417750de2a48d61

node scripts/docs-audit/affected-docs.mjs --json 601832b1a667992e73d54bf11643849836a6f778

⚠️ That checkout carried uncommitted changes, so the commit above does not fully identify what was read.

Advisory only, and a precision-first one (#9192): a page is listed because it names a
symbol, wire route or SDK method this diff touched — not because it mentions a changed
package. Each row says which anchor put it there, so a wrong row is reportable rather than
merely annoying. To re-verify, run the docs-accuracy-audit workflow scoped to these files:
node scripts/docs-audit/affected-docs.mjs 601832b1a667992e73d54bf11643849836a6f778 → pass the list as
args.docs, on the commit named under Which tree this was computed on.

[objectstack-fleet]

domain:server#4 · os-warren · session_01JQwxbacSXJQqAztgt4UMrS

席位核对

ACCEPT · domain:server#4 · head ae528b82771c30d29c90b7370417750de2a48d61 · 2026-11-03T05:10Z · seat review of record at TIER_DEFAULT for the whole PR. Clause-②: no and Metadata-protocol: none stand (a move; the published index.d.ts is byte-identical), so no contract review is owed. Not governed: check-governed-merges --pr 29788 0 of 6 paths, +348/−275.

1. Shape: draft, base main, Part of #25467 (ruling E 6537939895, E2 section 3, scoped by seat ruling A 6594932448). 6 files: the bodies of validateData, createData, updateData and deleteData move to protocol-data-writes.ts and reach the class through protocol-host.ts (three host members added: assertObjectRegistered, assertVersionMatch, probeRecord); the class keeps each docblock and signature as a one-line stub; byIdUpdateRecord leaves the class's import with updateData; the double-contract seams ledger moves two rows to the new file (--write) and its self-test names the new path, the same rule judged; one patch changeset. Inside claim 6594815331 as widened by ruling A.
2. Premise measured: section 2 had already moved (PR refactor(metadata-protocol): move the query validation out of protocol.ts into protocol-query-validation.ts (#25467 cut 3) #25538), so section 3 is the cut. Of its eight bodies, four move; findData, getData, cloneData and probeRecord stay because their six erasure sites are grandfathered by file in the shrink-only baseline (follow-up metadata-protocol: type the six query-options erasures in protocol.ts data reads (findData, getData, cloneData, probeRecord) so their bodies can leave the class #29764). The host-split check is text-identical for all four, and its negative control bites (dev report 6595748296).
3. Gates on this head: check-runs pinned to ae528b82: 48 runs, 35 success, 13 skipped, none failed. Dev local on the merged head (over refactor(metadata-protocol): getMetaTypes() declares a named return type, pinned equal to the type it returned #29730 and fix(metadata-protocol): a flow read's _diagnostics and the restore doors judge a stored flow in its D2-converted spelling #29753): metadata-protocol 310 files / 22,301 tests passed; dispatch-gates --ran 74 derived / 73 run / 0 UNRUN / 1 NOT MEASURED (dual-build-cjs-loads, prerequisite; CI covers it); pnpm lint 0; the erasure ratchet holds; index.d.ts byte-identical to main (ESM and CJS).
4. Accepted deviations: the cut is about 300 lines, under ruling E's 1,000-line guide, on the seat's ruling A; the two gate files are path-only moves.
5. Out of scope: metadata-protocol: type the six query-options erasures in protocol.ts data reads (findData, getData, cloneData, probeRecord) so their bodies can leave the class #29764 types the six erasures, then the last four bodies move as one more E2 cut.
6. Landing: not governed, under 3000 lines, no contract review owed ⇒ the seat lands it (pr_ready + automerge_enable). At MERGED: progress record on [maintainer] refactor(metadata-protocol): split src/protocol.ts (33,065 lines, +10,169 in 30 days) along its ObjectStackProtocolImplementation seams, one move-only slice per PR #25467 (stays open), the protocol.ts freeze lift note on seat post [PM seat] domain:server · seat 4 — ⏳ vacant #27737, pm:dispatched → pm:queue with the next step named.