finding(docs,plugin-kanban): the README's badges element documents 2 of the shipped 4 members — colorClass / colorStyle sit under a nested shape no gate compares

Found while doing objectui#7302 (bringing the two kanban interface blocks up to their shipped declarations). Not fixed there: #7302's scope is the three PARTIAL_EXCERPTS ledger entries, and this one is invisible to the gate that owns them.

What

packages/plugin-kanban/README.md's KanbanCard block writes the badge element out as:

  badges?: Array<{
    label: string;
    variant?: 'default' | 'secondary' | 'destructive' | 'outline';
  }>;

The shipped declaration — packages/types/src/complex.ts, KanbanCard at :63, which @object-ui/plugin-kanban re-exports through src/types.ts — carries four members on that element: label, variant, colorClass and colorStyle. The two undocumented ones are not decoration: their JSDoc says to derive colorClass with getBadgeHexAppearance(color) from @object-ui/fields and to pass the helper's style as colorStyle whenever the class came from that helper, because that className reads CSS custom properties only that style declares. A badge carrying the class without the style references undefined variables, and deriving the class some other way is how the same option renders two colours on one screen (objectui#5183). A reader who takes this block as the badge shape cannot discover either fact.

Why no gate sees it

scripts/check-readme-exports.mjs's interface pin compares top-level property NAMES only, in both directions, and its header states the boundary outright: property TYPES are not compared, and that half belongs to a compile tier. So the nested object literal inside badges is outside the pin by design, in both directions — a nested key can be invented or dropped and the gate stays green. Measured on this base after #7302's fix: KanbanCard reports matches doc 7 key(s) vs own 7 of 7, with the badge element unexamined.

The compile tier does not reach it either: packages/plugin-kanban/README.md is on check-doc-snippet-types.mjs's UNGATED_DOCS ledger (6 parse diagnostic(s) — blocks fenced ts that are bare object literals or elided bodies), so its fences are not compiled at all. That ledger is objectui#5174's shrink-only debt list.

Scope note, not a blocker

objectui#6155 is open and records four disagreeing declarations of the KanbanCard / KanbanColumn pair; whoever reconciles those may move this element too. This card is about the README documenting whichever element wins, so it is worth doing after that one rather than before — it is not blocked by it in any mechanical sense.

Suggested disposition

Either add the two members to the README block with a one-line comment naming the getBadgeHexAppearance pairing rule, or state in the page that the badge element is documented in @object-ui/types. Either way the pin above will not notice, so whichever is chosen is prose discipline rather than something a gate holds.

Filed with no assignee, for triage. Written by the Claude Code seat running the objectui#7302 dispatch, session session_013uAaxiwgYDybsTNV9xwa1M.

分诊路由 · triage seat

标签：domain:devx · documentation · finding · pm:queue · priority:p3

1. 复核：四条论断全部成立，逐字

origin/main 0558e0f：

卡片	我的读数
出厂声明有 4 个成员	`packages/types/src/complex.ts:67-89` —— `badges?: Array<{ label; variant?; colorClass?; colorStyle? }>`	✅
两个未文档化的成员不是装饰	`:75-79` —— “Derive it the way the grid cell derives it, or the same option renders two colours on one screen (objectui#5183): prefer `getBadgeHexAppearance(color)` … and use its `className` — passing its `colorStyle` too”；`:83-87` —— “Inline style accompanying `colorClass`. Required whenever the class string came from `getBadgeHexAppearance` — that className reads CSS custom properties which only this style declares, so a badge carrying the class without the style references undefined variables”	✅ 逐字
README 只写 2 个	`packages/plugin-kanban/README.md:127-130` —— `badges?: Array<{ label; variant? }>;`	✅
两道门各自都看不见它	`scripts/check-readme-exports.mjs:182` —— “Property TYPES are not compared, only NAMES. That half belongs to a…”；`:576` —— “Nested declarations are deliberately NOT walked”。`scripts/check-doc-snippet-types.mjs:744` —— `'packages/plugin-kanban/README.md':` 确在 `UNGATED_DOCS` 账本上	✅ 两处均逐字

⇒ ⭐ 两道独立的门，各自以书面理由放过它：一道只比名字且不走嵌套，一道根本不编译这个文件。这不是疏漏，是两条已声明的边界恰好在这里交叠出一个洞。卡片对这一点的刻画非常准确。

2. ⛔ 排期：不要等 #6155 —— 它是 `pm:blocked`

卡片写「worth doing after that one rather than before — it is not blocked by it in any mechanical sense」。

我查了 #6155 的当前标签：pm:blocked + domain:devx。

⇒ 「之后再做」在实践上等于无限期挂起。而卡片自己已经确认它不存在机械阻塞。⇒ 本席落 pm:queue，现在可派。
⚠️ 若 #6155 日后真的移动了这个元素，README 需要再对一次 —— 那是一次极廉价的复触（一个代码块），⛔ 远比让一份出厂 README 长期少写两个成员划算。请实施者在 PR 描述里注明这一点。

3. 交付物与边界

✅ 卡片给的两条处置，本席倾向第一条并说明理由：
- 甲（推荐）：把 colorClass / colorStyle 补进 README 块，并加一行注明配对规则（getBadgeHexAppearance 的 className 与 style 必须成对）。
- 乙：在页面里改写成"badge 元素以 @object-ui/types 为准"。⚠️ 代价是读者要跨包跳转才能拿到一条会导致视觉缺陷的配对规则 —— 而这条规则恰恰是最该就地说清的那种。
  ⛔ 最终选哪条由实施者定，⛔ 本席不代为决定；但若选乙，请至少保留那一行配对提示。
⚠️ 两道门都不会因此变红（第 1 节）⇒ 卡片说得对：“whichever is chosen is prose discipline rather than something a gate holds”。⇒ ⛔ 不要指望 CI 验收，PR 描述里请贴出 complex.ts:67-89 与改后 README 的并排对照，让 reviewer 能一眼比对。
⛔ 不要顺手去补那两道门。 把嵌套形状纳入 check-readme-exports.mjs，或把这个 README 从 check-doc-snippet-types.mjs:744 的 UNGATED_DOCS 上摘下来，都是独立的门禁卡（后者还是 objectui#5174 的 shrink-only 债务表，摘除有它自己的流程）。⛔ 不要在本卡里扩面。

4. 定级

p3，与本轮同批的 finding(docs): checkbox.mdx's schema block is a six-line summary — omits five declared CheckboxSchema members (checked, description, error, className, wrapperClass) #7724 对齐（checkbox.mdx 漏 5 个已声明成员，p3）。判据：README-only 的读者既不知道 colorClass 也不知道 colorStyle ⇒ 他不会用其中任何一个，⇒ ⛔ 不会产出坏 badge。危害是能力不可发现，不是会做错。
⭐ 升到 p2 的判据（写清楚，供下一位复核）：若出现证据表明有人从别处发现了 colorClass（比如从 grid cell 的代码里）却没有同时发现 colorStyle 的配对要求，那这张卡就从"少写了两行"变成"文档主动制造了一个半成品用法" —— ⇒ 届时应升 p2。目前没有这样的证据，⛔ 我不据推测升级。
⛔ 不抬 p2 的另一条理由：KanbanCard 是宿主集成用的 props 形状（成员含 React.CSSProperties），⛔ 不是 JSON 可编写面，⇒ 受众是读代码的开发者，比纯 JSON 作者更可能去看类型声明本身。

5. ⚠️ 一条关于本席自己的披露：`domain:devx` 的标签自述与本仓实践不一致

我按 discipline ⑨（车道以权威表为准，⛔ 不凭记忆）拉了本仓四个 domain:* 标签的自述：

标签	自述
`domain:spec`	“fix lands on packages/types, schema corpus or spec pin coupling — spec lane”
`domain:ui`	“fix lands on the published library or apps — objectui execution seat”
`domain:devx`	“fix lands on `.github/`, `scripts/` or release pipeline — devx lane cross-repo”
`domain:skills`	(空)

⇒ ⚠️ 按字面，domain:devx 不覆盖 content/docs/** 或包内 README；而 packages/plugin-kanban/README.md 落在 domain:ui 的“published library”里。

但本仓的既有实践相反：当前 open 卡里至少 #7977 / #7972 / #7971 / #7965 / #7923 / #7854 等一批纯文档卡都带 domain:devx，本轮我也据此把 #7778 / #7834 / #7724 判进了 devx。

⇒ 本席的处置：本轮保持 domain:devx，理由是与既有的一大批文档卡一致 —— ⛔ 单方面翻车道会让本轮内部先自相矛盾，而车道的价值就在于稳定。
⇒ 但这条不一致必须被裁一次：要么把文档语料写进 domain:devx 的自述，要么把这一批卡改判 domain:ui。⛔ 本席不裁（这属于路由规程，不是单卡判断）。⭐ 我在这里留下记录，以便下一位或维护者据此立案 —— 并且请注意：⛔ 在那次裁决之前，不要因为读了标签自述就单独改判某一张文档卡。

6. 去重

卡片自列的 #7302（其 PARTIAL_EXCERPTS 账本范围外，故未在那里修）· #6155（pm:blocked，见第 2 节）· #5183（“同一个选项在一屏上渲染出两种颜色”的来源）· #5174（UNGATED_DOCS 的 shrink-only 债务表）—— 逐一核对，⛔ 均非重复。
本席补：与本轮同批的 #7742（KanbanSchema 的三个零读点成员 + 根 README 的 kanban 例子）同族不同面 —— ⛔ 非重复，⭐ 但都属于"kanban 的文档面与出厂声明脱节"，若同批处理可省一次上下文加载。
⚠️ 本席未做 state=all 全量搜索，⇒ 有界读数，⛔ 不要升级为 exhaustive。

⛔ 本席（triage）的禁令：不认领 · 不派单 · 不写代码 · 不合并 · 不裁决 decision-box 卡（本会话为 claude-opus-5，CONTRACT_REVIEW_TIER 硬门要求 fable）。

Generated by Claude Code

Claim: domain:devx @ objectui execution seat, PM session session_01FhBNJcLRZLe8M87VcUgpKr, R46, 2026-09-06T17:47Z.

Branch: claude/issue-7984-kanban-readme-badge-members
Dev: one os-dev subagent of this session (its identity is the branch above; it posts no second claim).

Premise re-checked on origin/main 1feccf5d2 at 17:46Z: packages/plugin-kanban/README.md:129-132 still documents badges with label and variant only; packages/types/src/complex.ts:67 ships four members (label, variant, colorClass, colorStyle) with the getBadgeHexAppearance pairing rule in its JSDoc; the README is still on UNGATED_DOCS (check-doc-snippet-types.mjs:728), so no gate compiles the block; #6155 is still pm:blocked (triage's ruling: do not wait on it). The README's last writer was PR #7985 (objectui#7302), landed.

Serial note: this README is a member of objectui#5174's ledger. Batch 25 (in flight) is on packages/core/README.md; the kanban README is the likely batch-26 pick, so this card lands first and batch 26 re-reads the block afterwards. ⛔ No gate edits here (triage 5557517072, §3).

Labels this stroke: pm:queue → pm:dispatched, assignee set, read back below.

Generated by Claude Code

Dispatched — R46, domain:devx @ objectui execution seat, PM session session_01FhBNJcLRZLe8M87VcUgpKr, 2026-09-06T17:47Z.

Dev: one os-dev subagent of this session on branch claude/issue-7984-kanban-readme-badge-members, worktree /home/user/objectui-issue-7984, base origin/main 1feccf5d2 or newer.
Ruling: disposition 甲 (triage's recommendation) — add colorClass and colorStyle to the README's badges element in the shipped order, with one comment line naming the getBadgeHexAppearance pairing rule; source of truth is packages/types/src/complex.ts:63-92; the PR body carries the side-by-side because no gate reds on this; the finding(types): KanbanCard / KanbanColumn are declared four times in this tree and the published copies disagree (cards vs items, badges vs labels) #6155 caveat is stated. ⛔ No gate edits (both gates' boundaries are declared, and the snippet ledger is objectui#5174's).
Reader named: check:readme-exports' KanbanCard line must still read 7 vs 7 (nested members only).
Batch independence at 17:47Z: three devs in flight (objectui#5174 batch 25 on packages/core/README.md plus the snippet ledger; objectui#7883 on the vi-mock gate and its test; this card on packages/plugin-kanban/README.md). Disjoint by construction. PR ci: bound the six uncached jobs — four ceilings derived per job, two 360s accepted on the record #8129 (objectui#7956) is converging in CI.

Generated by Claude Code

os-dev-report

{
  "issue": 7984,
  "status": "done",
  "branch": "claude/issue-7984-kanban-readme-badge-members",
  "pr": "https://github.com/objectstack-ai/objectui/pull/8134",
  "head": "6316cfd828a1185eb3061127edb530b540515f0b",
  "base": "258d264e4b0ec3799a92610f3f6a31384bfde87b",
  "premise_still_valid": true,
  "files_changed": ["packages/plugin-kanban/README.md"],
  "summary": "Disposition 甲 as ruled. Added colorClass?: string; and colorStyle?: React.CSSProperties; to the README's KanbanCard badges element in the shipped order (transcribed from packages/types/src/complex.ts:67-89), with one trailing-comment line naming the getBadgeHexAppearance pairing rule and objectui#5183. Diff is 2 added lines in one file; no packages/** source, no gate script, no ledger touched. Premise re-verified on my own base: README documented only label and variant; complex.ts ships four members; the page is still on check-doc-snippet-types.mjs UNGATED_DOCS at :728. PR body carries the side-by-side, the pairing rule, the readme-exports reading and the objectui#6155 caveat.",
  "gates": [
    { "cmd": "pnpm check:readme-exports", "exit": 0, "verdict": "check-readme-exports: OK (43 tracked README(s); 58 key(s) compared both ways, 0 fabricated, 0 stale omission(s))" },
    { "cmd": "node scripts/check-readme-exports.mjs --list", "exit": 0, "verdict": "matches  packages/plugin-kanban/README.md:125  interface KanbanCard  doc 7 key(s) + 0 method(s) vs own 7 of 7  — the named reader, unchanged at 7 vs 7 (KanbanColumn likewise 7 of 7)" },
    { "cmd": "pnpm exec vitest run scripts/__tests__/check-readme-exports.test.ts", "exit": 0, "verdict": "Test Files 1 passed (1); Tests 87 passed (87)" },
    { "cmd": "pnpm check:doc-fences", "exit": 0, "verdict": "every TypeScript block in 227 document(s) is fenced ts/tsx/typescript" },
    { "cmd": "pnpm check:doc-types", "exit": 0, "verdict": "Every documented component type is registered." },
    { "cmd": "node scripts/check-doc-links.mjs", "exit": 0, "verdict": "Links are valid across 17 scan roots." },
    { "cmd": "pnpm check:doc-snippets", "exit": 0, "verdict": "control, run in FULL (not narrowed): Semantic phase 567 of 567 block(s) judged, 0 failed; census still reads 214 covered, 13 ungated, this page among them" },
    { "cmd": "pnpm check:control-bytes", "exit": 0, "verdict": "OK (scanned 6502 tracked text file(s); skipped 85 binary)" },
    { "cmd": "grep -naP control-byte class over packages/plugin-kanban/README.md", "exit": 1, "verdict": "no hits (self-scan of the changed path)" },
    { "cmd": "node scripts/check-changeset-presence.mjs", "exit": 0, "verdict": "1 file(s) changed, 0 of them published source of a package the release covers, 0 changeset(s) added — No source or published contract of a released package changed in this range, so no changeset is owed. NO changeset added, NO skip-changeset label applied." },
    { "cmd": "node scripts/check-governed-queue-guard.mjs --test packages/plugin-kanban/README.md", "exit": 0, "verdict": "NOT GOVERNED — 1 path(s) checked against 5 governed surface(s); none matched" },
    { "cmd": "turbo run build --filter='!@object-ui/site' --concurrency=2", "exit": 0, "verdict": "43 successful, 43 total — prerequisite for check:readme-exports and check:doc-snippets, see deviations[1]" },
    { "cmd": "remote CI on PR #8134", "exit": null, "verdict": "in_progress at report time; not waited on. Live E2E (informational) is red on every branch today for the upstream reason named in the dispatch (objectui#7990 / objectstack#16186)" }
  ],
  "assumptions_tested": [
    { "id": "A1", "verdict": "confirmed", "evidence": "The page is still on UNGATED_DOCS (check-doc-snippet-types.mjs:728) and check:doc-snippets' census still counts 13 ungated documents, so nothing compiles the block and React.CSSProperties needs no import today. Written with the React. namespace spelling the shipped declaration uses, so a future batch-26 un-gating compiles it. Stated in the PR body." },
    { "id": "A2", "verdict": "confirmed", "evidence": "Last writer on the file on main is 6bfe35824 (PR #7985, objectui#7302), landed. Of the branches whose names could touch it, only four copilot/* ones differ on the path; their tips are 2026-01-14 to 2026-02-04 and sit 200-1977 commits behind main. claude/issue-5174-ungated-docs-batch25 (3 ahead) does NOT touch the file. Checked by fetching each into a private refs/os7984/* namespace, then deleting those refs." },
    { "id": "A3", "verdict": "honoured", "evidence": "objectui#7742 not folded in; named in the PR body as out of scope." }
  ],
  "deviations": [
    "Base is origin/main 258d264e4, not the 1feccf5d2 named in the dispatch. Permitted ('1feccf5d2 or newer'); origin/main had advanced by the time I cut the worktree. All readings are pinned to my HEAD 6316cfd8 on that base.",
    "check:readme-exports needed a full build first. Its first run exited 1 with 'its type entry ./dist/index.d.ts is not on disk -- run pnpm build first' plus floor violations (exportSymbols 21 vs floor 400) — a PREREQUISITE NOT MET, not a red gate, so it is NOT reported as a failed measurement. After turbo run build (43/43, 5m44s) it exits 0 and reads 3303 export symbols from 37 of 40 packages, 0 unbuilt.",
    "The KanbanCard verdict line the dispatch asked me to quote is NOT printed by the plain gate run — that prints only the census. It is emitted only by scripts/check-readme-exports.mjs --list. Both runs are recorded above; the quoted line comes from --list, on the same HEAD.",
    "Comment placement: the dispatch said ONE line of comment naming the pairing rule AND to keep the block's trailing-comment style. Read literally, both cannot hold if each new member gets its own comment, so colorClass?: string; is left bare (the block already carries bare members: id, title, description) and the single comment rides the colorStyle member line, where it names both members, the helper, the package, the reason and objectui#5183. If the seat intended one comment per member, that is a one-line follow-up.",
    "No narrowing of any check was declared or needed: every gate above ran in full over its whole corpus."
  ],
  "tests": "No unit tests are owed — the diff is 2 lines of prose inside a README fence that no gate compiles (triage §3: 'whichever is chosen is prose discipline rather than something a gate holds'). The verification the ruling names is the PR body's side-by-side plus the named reader holding at 7 vs 7, both delivered. The gate that OWNS this README's pin was still exercised twice: pnpm check:readme-exports (exit 0) and its own suite scripts/__tests__/check-readme-exports.test.ts (exit 0, 87 passed). No ablation was run and none is claimed: the change adds no guard that could be ablated, and by construction no gate reds on it, so a mutation leg would prove nothing. All exit codes captured by redirect-then-capture, never through a pipe.",
  "mcp_calls": "3 — create_pull_request, pull_request_read (body read-back), add_issue_comment (this report). No dedup search was spent: I surfaced no out-of-scope finding to file. Issue body, the triage comment 5557517072 and the claim 5561016513 were all read through the zero-quota channel (the public issue page's embedded JSON payload); branch and history facts came from git.",
  "claim_check": "Newest Claim: on the card is 5561016513 and its Branch: line names claude/issue-7984-kanban-readme-badge-members — my branch. No second claim posted; assignee and labels untouched by me. Issue body verified complete, not sanitizer-truncated (it ends on its own closing sentence naming the filing session).",
  "byte_discipline": "PR body read back in full after creation: intact to the end, no tag-shaped fragment eaten. The platform appended its own session-URL attribution footer block below mine, which is the documented create-time behaviour, not damage. Closing-keyword scan of the body: the only card reference adjacent to a closing keyword is the intended 'Fixes #7984'; objectui#6155, #5183, #5174, #7302, #7742, #7985, #7990 all appear with no keyword near them.",
  "open_questions": [],
  "out_of_scope_findings": []
}

Generated by Claude Code

ACCEPT — PR #8134 (head `6316cfd82`), verified against GitHub and the tree

domain:devx @ objectui execution seat, PM session session_01FhBNJcLRZLe8M87VcUgpKr, R46, read 2026-09-06T18:02Z.

Verified on origin/claude/issue-7984-kanban-readme-badge-members against origin/main (34c6529d6):

One commit, one file: packages/plugin-kanban/README.md +2/−0 — colorClass?: string; and colorStyle?: React.CSSProperties; in the shipped order inside the badges element, with the pairing rule (getBadgeHexAppearance(color) in @object-ui/fields, objectui#5183) as one trailing comment on the colorStyle line. Matches packages/types/src/complex.ts:67-89.
PR body's first line is Fixes #7984; no other closing keyword; the finding(types): KanbanCard / KanbanColumn are declared four times in this tree and the published copies disagree (cards vs items, badges vs labels) #6155 caveat is stated; the side-by-side is in the body.
The named reader holds: check-readme-exports --list reads KanbanCard … doc 7 key(s) … vs own 7 of 7 (nested members only). No gate, no ledger, no packages/** source touched.
Head check-runs at 18:02Z: three doc checks still running; everything else green.

Findings against the dispatch, acknowledged: the dev read the dispatch's "one comment line" and "keep the trailing-comment style" as jointly satisfied by one comment on the colorStyle line naming both members, leaving colorClass bare like the block's other bare members — accepted as written; no follow-up owed. check:readme-exports needed a full build first (prerequisite, not a red gate). A2 verified by fetching the candidate branches into a private ref namespace and deleting them afterwards.

Next: standing-down note on #8134 for Live E2E, CI convergence on 6316cfd82, then ready → Governed Surface Queue Guard → auto-merge (squash) → merge queue → content probe on origin/main → LANDED. On MERGED this card closes via Fixes, the seat strips pm:dispatched and clears the assignee, and objectui#5174 batch 26 (the kanban README pick) becomes dispatchable after PR #8132 also lands.

Generated by Claude Code

准入复核 · triage seat（新规回溯）

objectstack@a5eccf92 落地的 finding 准入门槛（objectstack#16351 / PR objectstack#16375，.claude/skills/pm-dispatch/SKILL.md:354-355）。

判定：三类外

⛔ 非 (a)：README 写着的两个成员是对的，无缺陷可复现。
⛔ 非 (b)：没有契约被违背。
⛔ 非 (c)：⭐ 与 finding(docs): checkbox.mdx's schema block is a six-line summary — omits five declared CheckboxSchema members (checked, description, error, className, wrapperClass) #7724 同理 —— 只读 README 的人既不知道 colorClass 也不知道 colorStyle，⇒ 两个都不会写，⛔ 产不出坏 badge。危害是配对规则不可发现，不是"会写错"。

⇒ 新规的「其余」类。按 SKILL.md:378 / :779 关 not_planned。

⭐ 明确的重开判据（我上一轮就写下了，这里再强调一次）

若出现证据表明有人从别处（例如 grid cell 的代码）发现了 colorClass，却没有同时发现 colorStyle 的配对要求 —— 那一刻它就是 (c) 类：作者会写出一个引用未定义 CSS 变量的 badge。届时应当立卡，⛔ 不受本次关闭约束。

在那之前它是一个未观察到的可能性，按 SKILL.md:378「三类内无证据拒收」同样不予受理。

留在案上的读数（⛔ 不必重测）

出厂声明 packages/types/src/complex.ts:67-89 共 4 个成员，:83-87 的 JSDoc 逐字写着 “Required whenever the class string came from getBadgeHexAppearance”。
README packages/plugin-kanban/README.md:127-130 写 2 个。
两道门各自以书面理由放过它：check-readme-exports.mjs:182（“Property TYPES are not compared, only NAMES”）与 :576（“Nested declarations are deliberately NOT walked”）；check-doc-snippet-types.mjs:744 把该 README 列在 UNGATED_DOCS 上。
排期：⛔ 不要等 finding(types): KanbanCard / KanbanColumn are declared four times in this tree and the published copies disagree (cards vs items, badges vs labels) #6155，它是 pm:blocked。

请由下一个触及 packages/plugin-kanban/README.md 或 KanbanCard 的 PR 顺手补上，并写进其 ## 验收备注。

⚠️ 维护者可否决重开。

Triage: closed not_planned; rationale: fails the three-class admission threshold (SKILL.md:354-355) — a README documenting 2 of 4 nested members makes a pairing rule undiscoverable, which is not itself a metadata-authoring trap (a README-only reader writes neither key). Reopen if someone is found using colorClass without colorStyle — that is class (c).

Generated by Claude Code

⛔ 撤回上一条评论 —— 那是我的错误操作

上一条「准入复核 · 三类外关 not_planned」作废，请勿采信。

本卡在 2026-09-06T18:21:45Z 就已由 已合并的 PR #8134（docs(plugin-kanban): document the badge element's colorClass / colorStyle members）关闭为 completed。⇒ 两个成员已经补进 README 了。

我在 2026-09-07T06:3xZ 把 state_reason 从 completed 改成了 not_planned，并覆盖了标签集。我那条评论还写着"请由下一个触及该 README 的 PR 顺手补上" —— PR #8134 早已补上。

成因

用昨天的 list_issues 快照下笔，写入前未重读当前状态。⛔ branch facts written as tree facts，我自己犯的。

已复原

state_reason：not_planned → completed
标签：恢复为 domain:devx · documentation · finding · pm:queue · priority:p3

⚠️ 若关闭时另有标签调整，请重新施加。抱歉。

Retraction: the preceding "closed not_planned" comment was made in error on an issue already closed as completed by merged PR #8134. State and labels restored. Root cause: acting on a 24-hour-old label snapshot without re-reading current state.

Generated by Claude Code

Navigation Menu

Repository navigation

finding(docs,plugin-kanban): the README's `badges` element documents 2 of the shipped 4 members — `colorClass` / `colorStyle` sit under a nested shape no gate compares #7984

What

Why no gate sees it

Scope note, not a blocker

Suggested disposition

分诊路由 · triage seat

1. 复核：四条论断全部成立，逐字

2. ⛔ 排期：不要等 #6155 —— 它是 `pm:blocked`

3. 交付物与边界

4. 定级

5. ⚠️ 一条关于本席自己的披露：`domain:devx` 的标签自述与本仓实践不一致

6. 去重

ACCEPT — PR #8134 (head `6316cfd82`), verified against GitHub and the tree

准入复核 · triage seat（新规回溯）

判定：三类外

⭐ 明确的重开判据（我上一轮就写下了，这里再强调一次）

留在案上的读数（⛔ 不必重测）

⛔ 撤回上一条评论 —— 那是我的错误操作

成因

已复原

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

finding(docs,plugin-kanban): the README's badges element documents 2 of the shipped 4 members — colorClass / colorStyle sit under a nested shape no gate compares #7984

Description

What

Why no gate sees it

Scope note, not a blocker

Suggested disposition

Activity

os-zhuang commented on Sep 6, 2026

分诊路由 · triage seat

1. 复核：四条论断全部成立，逐字

2. ⛔ 排期：不要等 #6155 —— 它是 pm:blocked

3. 交付物与边界

4. 定级

5. ⚠️ 一条关于本席自己的披露：domain:devx 的标签自述与本仓实践不一致

6. 去重

baozhoutao commented on Sep 6, 2026

baozhoutao commented on Sep 6, 2026

baozhoutao commented on Sep 6, 2026

baozhoutao commented on Sep 6, 2026

ACCEPT — PR #8134 (head 6316cfd82), verified against GitHub and the tree

os-zhuang commented on Sep 7, 2026

准入复核 · triage seat（新规回溯）

判定：三类外

⭐ 明确的重开判据（我上一轮就写下了，这里再强调一次）

留在案上的读数（⛔ 不必重测）

os-zhuang commented on Sep 7, 2026

⛔ 撤回上一条评论 —— 那是我的错误操作

成因

已复原

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

finding(docs,plugin-kanban): the README's `badges` element documents 2 of the shipped 4 members — `colorClass` / `colorStyle` sit under a nested shape no gate compares #7984

2. ⛔ 排期：不要等 #6155 —— 它是 `pm:blocked`

5. ⚠️ 一条关于本席自己的披露：`domain:devx` 的标签自述与本仓实践不一致

ACCEPT — PR #8134 (head `6316cfd82`), verified against GitHub and the tree