[objectstack-fleet]

Requested by Warren · project thread
Fixes #29485

Clause-②: no
Metadata-protocol: none

维护者速读

• 改了什么：email.defaultTemplateContext 的 .describe() 与 TSDoc 把 baseUrl 和 appName 并列为运行时自行解析的键，并写出 baseUrl 的解析链；重新生成 email-config.mdx。
• 为什么改：feat(plugin-email): every email template can read the deployment address as {{baseUrl}} #29478 已让邮件插件解析 baseUrl，规范文字仍说只有 appName 例外，读 /meta/types 或参考页的作者会被误导。
• 风险与回滚：只改文字，无 schema 变化，键仍是自由 record；回滚即 revert 本 PR。
• 先看：packages/spec/src/system/email-config.zod.ts（defaultTemplateContext 的文档块与 describe）。
• 两分钟验证：pnpm --filter @objectstack/spec build && pnpm --filter @objectstack/spec check:generated，预期 14 个产物全部最新。
• 需要你做的：无，席位落地。

What changes

• packages/spec/src/system/email-config.zod.ts: the defaultTemplateContext docblock and .describe() now name baseUrl beside appName as the two keys the runtime resolves itself. The baseUrl chain is read from the merged plugin code (resolveDeploymentBaseUrl in packages/plugins/plugin-email/src/capability-arg.ts, landed by feat(plugin-email): every email template can read the deployment address as {{baseUrl}} #29478 at 73c24394e2): OS_AUTH_URL (legacy BETTER_AUTH_URL), then OS_BASE_URL, then defaultTemplateContext.baseUrl, then http://localhost: plus the serve port (OS_PORT, legacy PORT, else 3000), with any trailing slash trimmed. The docblock also records that a send's own data wins over both keys (EmailService spreads the send's data over this context).
• content/docs/references/system/email-config.mdx: regenerated with pnpm --filter @objectstack/spec gen:docs (two table rows: the service schema and the stack email schema).
• .changeset/29485-email-template-baseurl-describe.md: @objectstack/spec patch. packages/spec ships src/**/*.zod.ts as source, so the docblock publishes.

No schema change: the key stays z.record(z.string(), z.unknown()).optional(), as the card requires.

Why

Ruling A on #29095 kept the spec out of that card and left this text to the spec lane. The describe still read "Passed through unchanged except appName", which is now incomplete.

Tests

No test pinned the old wording (git grep over the tree for the old describe and docblock sentences: only the generated mdx hit). Local runs on head cc250df8:

• pnpm --filter @objectstack/spec typecheck: exit 0.
• pnpm --filter @objectstack/spec exec vitest run --maxWorkers=2 src/system: 66 files, 2096 tests passed.
• pnpm --filter @objectstack/spec check:generated: all 14 generated artifacts up to date after gen:docs.

Acceptance notes

• The file-top docblock of email-config.zod.ts still says "appName is the one key whose env layer is not OS_EMAIL_*". That sentence is about the schema's own keys, and baseUrl is a key inside defaultTemplateContext, not a schema key, so it stays true and is outside the claimed file surface. Noted, not filed.

Verification

node scripts/pm/dispatch-gates.mjs --commands derived 97 gates for this diff on cc250df8; all were run and reconciled with --ran: 96 exit 0, 1 NOT MEASURED.

• NOT MEASURED: pnpm check:dual-build-cjs-loads, reason: exit 3, it needs every package built; left to CI.
• node scripts/check-plugin-teardown-shape.mjs --self-test first exited 3 (pinned fixture commit not in the shallow clone); after fetching that commit it exited 0 (48 cases).

Human Minimum

• 1. Schemas start as Zod; packages/spec holds no runtime logic.
• 2. Every name I added uses the casing for its kind.
• 3. No formatter ran; pnpm lint passes. Not run locally: the repo-wide lint run is CI's.
• 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: its generated artifacts are regenerated and committed.
• 7. No hand edits to content/docs/references/, packages/*/CHANGELOG.md or content/docs/releases/.
• 8. Title, body and comments are in English.

---

Generated by Claude Code

[github-actions]

📓 Docs Drift Check

1 anchor(s) derived from 1 changed package(s); no hand-written page names any of them, so this run has nothing to list — not a clean bill of health. This check sees only pages that NAME a derived anchor: one that documents this change in prose, or enumerates it in an authoring dialect, names none and stays invisible to it on every run.

What this run could not see

• the SDK route bridge reached 60 of 216 client-bound route-ledger rows — the other 156 have no registrar path: tail to select them, so pages documenting THEIR client methods cannot appear above, on this or any run. Of those 156: 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; 101 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 — 168 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 5651647d82dea5ce18ad537d629c18e3dd27f613 → packageMentionDocs.

Which tree this was computed on

This run read content/docs from f41a424c8f36baaab1b27db770c8a4ce843e83b8 — the merge of head cc250df8a0f36a33c470d073ffcaaa716f5b2118 into base 5651647d82dea5ce18ad537d629c18e3dd27f613, 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 f41a424c8f36baaab1b27db770c8a4ce843e83b8 && git checkout f41a424c8f36baaab1b27db770c8a4ce843e83b8
# afterwards, rebuild it from the two parents, which stay fetchable
git fetch origin 5651647d82dea5ce18ad537d629c18e3dd27f613 cc250df8a0f36a33c470d073ffcaaa716f5b2118 && git checkout -B drift-repro 5651647d82dea5ce18ad537d629c18e3dd27f613 && git merge --no-ff cc250df8a0f36a33c470d073ffcaaa716f5b2118

node scripts/docs-audit/affected-docs.mjs --json 5651647d82dea5ce18ad537d629c18e3dd27f613

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

[objectstack-fleet]

domain:spec#5 · os-warren · session_01JQwxbWR3K3h7i1ytwycVg1

Contract-review-claim: session_01JQwxbWR3K3h7i1ytwycVg1 · head cc250df · 2026-11-02T18:38Z

The published-schema face (packages/spec/src/system/email-config.zod.ts) is reviewed at CONTRACT_REVIEW_TIER by an isolated at-tier reviewer the dispatching seat spawns; it is not the drafter and is fed only the card, the PR and the check-runs.

[objectstack-fleet]

domain:spec#5 · os-warren · session_01JQwxbWR3K3h7i1ytwycVg1

Contract review

Served-tier: CONTRACT_REVIEW_TIER
Head-sha: cc250df8a0f36a33c470d073ffcaaa716f5b2118
Local-runs: none
Scope: packages/spec/src/system/email-config.zod.ts + packages/plugins/plugin-email/src/capability-arg.ts, packages/plugins/plugin-email/src/email-service.ts, packages/plugins/plugin-email/src/email-plugin.ts, packages/cli/src/commands/serve.ts, packages/cli/src/commands/serve-email-base-url-parity.test.ts, packages/spec/src/stack.zod.ts

Inputs read: card #29485 (body and all three comments: triage, Claim, os-dev-report), PR #29539 (body, 3-file list, net diff against main at the head above), and the 65 check-runs on the head. Read-only: the cited code was read on a local clone at a recent origin/main (390bd2ae) that contains both the PR base 5651647d and the #29478 landing commit 73c24394e2.

① Derived judgments

• Accept set — unchanged, right. defaultTemplateContext stays z.record(z.string(), z.unknown()).optional() on EmailServiceConfigSchema; StackEmailConfigSchema inherits it through .omit({ apiKey, options }), so nothing an author can write is newly accepted or refused on either schema. The card's "⛔ No schema change" holds.
• Public surface — a .describe() string and a TSDoc block, right. packages/spec ships src/**/*.zod.ts as source, so both publish; /meta/types and the generated reference carry the new describe. The describe carries no tracker number (check:doc-authoring shape holds); #5448, PR #5498 and #29095 sit in TSDoc only, in the file's existing citation style.
• Generated output — right. content/docs/references/system/email-config.mdx changes exactly two rows (the service schema and the inherited stack row), matching the two schemas that carry the key. A grep of the PR head finds no residue of the old wording ("except appName, which", "One key is not passed through") anywhere in the tree: no test pinned it, and no other generated artifact embeds it.
• Gates — all green. On the head, the ten required contexts are success (Lint & Repo Gates, TypeScript Type Check, Test Core, Build Core, Governed Surface Queue Guard, Lint UI is skipped by its own path filter with Type Check UI / Test UI / Build & E2E UI / Live E2E UI success); all sixteen Test UI shards completed success. Check Changeset, Check PR Size, PR-body guards and Spec property liveness are success. No governed path is in the file list.

② Semver level

.changeset/29485-email-template-baseurl-describe.md declares '@objectstack/spec': patch — right. The diff publishes text only (a describe string and a docblock in a shipped source file), no shape, no type, no default. Clause-②: no is on both the PR body and the changeset, consistent with each other and with a diff that neither widens nor narrows what the spec accepts. Metadata-protocol: none is right: email config is operator config, not a metadata type.

③ Boundary flags

open_questions: [] — nothing to answer. The five declared deviations, each judged:

1. Model-free trailer pair on both commits (Claude-Session + Co-authored-by: Claude) instead of the harness's model-named line — read off git log; this is what AGENTS.md and the pre-push hook require. Accepted.
2. Branch claude/project-thread-5qav58, not claude/issue-29485-* — recorded in the Claim comment as the card-thread platform exception; the PR body carries Fixes #29485, so discoverability is through the card. Accepted.
3. PR assignee / reviewer requests skipped per the dispatch order — no effect on the contract. Accepted.
4. Human Minimum 3 left unticked (pnpm lint not run locally) — Lint · ESLint and Lint & Repo Gates are success on the head. Accepted.
5. WIP push before the spec build — the head carries both commits and the generated row matches the source. Accepted.

Out-of-scope note (not filed by the dev): the file-top docblock still says "appName is the one key whose env layer is not OS_EMAIL_*". Judged against the schema: that sentence enumerates the schema's own keys (provider, apiKey, defaultFrom, retries, persist, queueDelivery, options, appName, defaultTemplateContext), and baseUrl is a member of the defaultTemplateContext record, not a key of the schema, so it stays true. Not escalated; the card's wanted list scoped the edit to the defaultTemplateContext docblock and describe.

④ Design judgment

packages/spec/src/system/email-config.zod.ts — judged against the ruling it implements (card #29485, which carries ruling A on #29095 barring the spec edit from that card and handing the text to the spec lane) and its first consumer (resolveEmailCapabilityArg in packages/plugins/plugin-email/src/capability-arg.ts, the single reader of config.email). Every new sentence about live behaviour, tried against the tree:

• "OS_AUTH_URL (legacy BETTER_AUTH_URL) → OS_BASE_URL → this map's baseUrl → http://localhost: plus the serve port" — true, in that order: resolveDeploymentBaseUrl reads env.OS_AUTH_URL ?? env.BETTER_AUTH_URL ?? env.OS_BASE_URL ?? cfgEmail.defaultTemplateContext?.baseUrl ?? \http://localhost:${port}\``.
• "the serve port (OS_PORT, legacy PORT, else 3000)" — true: const port = env.OS_PORT ?? env.PORT ?? '3000' in the resolver, and serve.ts's --port default is readEnvWithDeprecation('OS_PORT', 'PORT') ?? '3000'; the parity test pins both orders (OS_PORT over PORT, legacy PORT when OS_PORT unset).
• "with any trailing slash trimmed" — true: .replace(/\/+$/, '') on the resolved value, and the parity test's OS_BASE_URL: 'https://base.example.com/' case expects the slash gone.
• "Both are always present in the delivered context" — true: the resolver builds { ...cfgTemplateContext, appName: resolveDeploymentAppName(...), baseUrl: resolveDeploymentBaseUrl(...) }, so both keys are set on every boot; email-plugin.ts passes this.options.defaultTemplateContext to the service unchanged.
• "A send's own data still wins over both, as it does over every key here" — true: EmailService.sendTemplate builds { ...(this.options.defaultTemplateContext || {}), ...(input.data || {}) }, the per-call data spread last.
• "Writing it here is the fallback for a deployment that sets neither env var" — true: it is the fourth operand, after the three env reads and before the localhost tail. (Nuance, not a finding: ?? means a set-but-empty env var is a value and does not fall through to the config rung, as the resolver's own docblock records; the spec text says "env first" and does not claim otherwise.)
• "baseUrl follows the same order (plugin-email: the email template context has no deployment base URL, so a template cannot render a clickable record link #29095)" — env wins, as for appName; the citation is right. 73c24394e2 is feat(plugin-email): every email template can read the deployment address as {{baseUrl}} (#29478) with Fixes #29095 in its body, and it is an ancestor of main, so the card's premise ("lands after PR feat(plugin-email): every email template can read the deployment address as {{baseUrl}} #29478 merges") is satisfied.
• The describe's short clause ("baseUrl takes OS_AUTH_URL, then OS_BASE_URL, then the value written here, then the local http://localhost default, with no trailing slash") is the card's wanted shape — env first, the written value, the local default, no trailing slash — and matches the chain; it omits the legacy BETTER_AUTH_URL alias and the port, which the TSDoc above it carries. Acceptable brevity for a one-clause describe, and it introduces no {{…}} (the docs generator's brace trap the appName describe already notes).
• The appName chain restated in the new bullet is unchanged and still matches resolveDeploymentAppName (OS_APP_NAME || cfgEmail.appName || defaultTemplateContext.appName || configAppName || 'ObjectStack').

Pre-existing, re-wrapped by the hunk but not new: "the CLI passes this object through to EmailServicePlugin as written". On main the reader is resolveEmailCapabilityArg in @objectstack/plugin-email, invoked through @objectstack/core's capability-composition by both os serve and bootStack; "the CLI" is the historical phrasing the file-top docblock already corrects. Outside the card's wanted list; not a finding.

Findings: none must-fix, none should-fix.

Implemented-by: claude/project-thread-5qav58
Reviewed-by: session_01JQwxbWR3K3h7i1ytwycVg1

VERDICT: PASS

[objectstack-fleet]

domain:spec#5 · os-warren · session_01JQwxbWR3K3h7i1ytwycVg1

维护者速读

• 改了什么:email.defaultTemplateContext 的 .describe() 与 TSDoc,把 baseUrl 与 appName 并列为运行时自行解析的两个键,写明 baseUrl 的取值链;再生成两行参考文档。
• 为什么改:feat(plugin-email): every email template can read the deployment address as {{baseUrl}} #29478 落地后邮件插件会为模板解析部署地址 baseUrl,旧文字仍说「除 appName 外原样透传」,与行为不符。
• 风险与回滚:纯文字(+45 / -23),schema 形状不变;回滚即 revert。
• 先看:packages/spec/src/system/email-config.zod.ts 的 defaultTemplateContext 一段。
• 两分钟验证:对照 packages/plugins/plugin-email/src/capability-arg.ts L118–121 的 resolveDeploymentBaseUrl,顺序与去尾斜杠一致。
• 需要你做的:无,席位落地。
• 席位意见:at-tier 合同审查 PASS(记录 6585309061),逐句对过解析器与 sendTemplate 的合并顺序;无 must-fix、无 should-fix。

席位核对

ACCEPT · domain:spec#5 · head cc250df8a0f36a33c470d073ffcaaa716f5b2118 · 2026-11-02T18:44Z

1. Shape: three files, all inside claim 6583821341's surface: email-config.zod.ts (the defaultTemplateContext docblock and describe only), the generated content/docs/references/system/email-config.mdx (two rows: the service schema and the inherited stack row), .changeset/29485-email-template-baseurl-describe.md (@objectstack/spec patch). Draft, base main, Fixes #29485.
2. Premise: feat(plugin-email): every email template can read the deployment address as {{baseUrl}} #29478 (Fixes #29095) is on main; the seat re-read resolveDeploymentBaseUrl: OS_AUTH_URL ?? BETTER_AUTH_URL ?? OS_BASE_URL ?? defaultTemplateContext.baseUrl ?? http://localhost:${OS_PORT ?? PORT ?? '3000'}, trailing slashes trimmed; sendTemplate spreads the send's data last. Every new sentence holds.
3. Gates: 65 check-runs on this head, all complete, zero failures; required contexts success, skipped ones are path- or opt-in-scheduled (Lint UI, Dogfood Verify CLI, UI Published Artifacts, Temporal Conformance, Dogfood Regression Gate shards, 18.x upgrade harness, Schedule gate, Packed-tarball smoke). Check Changeset success.
4. Contract review: packages/spec/src/** hit ⇒ owed; claim 6585215467, isolated at-tier reviewer, record 6585309061 PASS on this same head. Accepted deviations: the five the dev declared, each judged in that record's ③.
5. Out of scope: the file-top line "appName is the one key whose env layer is not OS_EMAIL_*" stays true (baseUrl is a record member, not a schema key). Skills-docs: n/a (Metadata-protocol: none, Clause-②: no).
6. Landing: not governed (check-governed-merges --pr 29539: 0 of 3 paths, 68 changed lines ≤ 3000). Seat lands it: pr_ready + automerge_enable.

[github-actions]

⛔ merge queue 构建失败 — 先分诊，再决定要不要重排

队列构建 41503806204 红了。队列跑的是全量套件（PR 侧 CI 只跑 affected 子集），
所以失败的测试可能在本 PR 没碰过的包里 —— 那不是重排能修的。每次盲目重排都会让排在后面的所有 PR 重建一轮。

分类：failure —— 按下面的日志分诊。

失败的 job（日志抽取，best effort）：

• Test Core (2/8) — 失败步骤: Run this shard's tests

  @objectstack/spec:test:repo:  FAIL   repo  scripts/strictness-ledger-doc.test.ts > the ledger and the generated counts agree > is checked in current — every shard on disk equals a fresh render, and no
    ↳ 失败原因: @objectstack/spec:test:repo: AssertionError: expected Map{ …(14) } to deeply equal Map{ …(14) }
  @objectstack/spec:test:repo:  FAIL   repo  scripts/strictness-ledger-doc.test.ts > the ledger and the generated counts agree > the read-time totals equal the sums of the shards' own rows
    ↳ 失败原因: @objectstack/spec:test:repo: AssertionError: expected 'untriaged: 1298 object site(s) across…' to be 'untriaged: 1297 object site(s) across…' // Object.is equality
  

↳ 失败原因 是判读的关键：超时（Test timed out in … / Hook timed out in …）多半是负载/时序，不是本 PR 的回归；
断言（AssertionError: …）才指向真实的行为改变。两者的 FAIL 行长得一模一样，只有这一行能区分。

⚠️ 断言这一侧有一类例外，判据是断言在测什么，不是它是不是 AssertionError。 断言的对象是产品行为（一个值、一个形状、一次拒收）⇒ 照上面读：真实的行为改变，去查，⛔ 不要重排掉；
断言的对象是这次实验自身的有效性前提（跑完的耗时、负载下的先后、任何只在时间预算内才成立的条件）⇒ 它跟超时是同一类，同样对负载敏感，重排一次是合法的判别手段。
识别是机械的：断言的消息或它比较的值本身点名了一段时长、一个时间戳、一个耗时计数。实测过的一对 —— AssertionError: SecurityPlugin.init() ran: expected false to be true 测的是产品行为（真回归）；
AssertionError: this run took over a second, so second-precision stamps could have differed too: expected 1006 to be less than 1000 测的是实验前提：它守护的那条不变式当时是绿的，同一个 head 原样重排一次即成功。
穿着 AssertionError 外衣的时间测量，仍然是时间测量。（⛔ 这只改「怎么读一次红」，不改「哪些测试可以重排」——后者由别处管。）

跨 PR 相同签名（24h，按失败测试文件聚合）：

• scripts/strictness-ledger-doc.test.ts — 24h 内已弹出 16 个 PR（feat(spec): declare the whole-deployment export, every installed package as installable packages in one read (PackageExportBundleSchema) #29100、feat(spec): push channel protocol face — device registration widening, stored device shape, device contract members, push route, APNs/FCM config by credentialRef #29218、fix(spec)!: refuse a dashboard globalFilters entry keyed dateRange beside a declared dateRange #29370、fix(spec)!: a datasource named drivers is refused at every door that parses a datasource #29423、feat(spec): a record page can name the apps whose routes open it (PageSchema.apps) #29437、docs(spec): tenancy-posture prose says which side of the open/commercial split gates activation #29454、feat(spec): declare features.sso on the auth config response #29455、feat(lint)!: os lint refuses a flow region the engine refuses to register (flow-region-shape-invalid) #29495、build(turbo): hash the Service Worker generator into the Console's build and type-check only #29504、fix(pm): the relay reads no runner 403 as the egress proxy's, and a refused advisory row names its cause #29505、fix(cli): os migrate organization-ownership --apply carries a public-form withdrawal onto a read-only package view into its .custom pair, declared #29508、docs(spec): the disabledActions description names Setup › Packaged Actions, not the retired Packaged automation page #29511、docs(spec): TOP_LEVEL_COLLECTIONS_NEVER_RESIDUAL names the readers of top-level datasets the tree has #29514、feat(pm): a ruled decision card still in the box is a patrol row, and the director's ruling lands with its knock #29515、feat(spec): the package export manifest declares withheldMaps, classified door for the unmanaged import #29521、docs(spec): email.defaultTemplateContext names baseUrl beside appName as a resolved key #29539），分属 2 条投机栈 ⇒ 2 次独立命中。汇总 issue：Queue-flake anchor: scripts/strictness-ledger-doc.test.ts #29524（已刷新）
• ⚠️ 24h 评论账本没读完（超过 5 页仍未读到窗口尽头），所以上面的「不同 PR 数」是下界，不是全量。

历史信号：

• 本 PR 过去 24h 无队列失败记录（首次）。
• 过去 24h 队列共有 33 个失败构建（不含本次）。

分诊清单：

1. 失败测试在本 PR 改动的包里 → 真回归，修 PR。
2. 失败测试与本 PR 无关 → 看上面的「跨 PR 相同签名」；已有汇总 issue ⇒ flaky/环境问题实锤，去那张 issue 上谈，修好前重排只会再烧一轮全队列。
3. 两者都不是 → 可能与同组 PR 语义冲突；等前面的 PR 落地或失败出队后再重排一次即可，不要连续重排。

---

Generated by Claude Code · merge-queue-triage workflow (#4859)