@topmindspace/topmind-skills 4.13.3 → 4.13.4

package/INSTALL.md

@@ -231,7 +231,7 @@ push tag `v*` 后 CI 自动：
 1. 测试 + 打包
 2. 创建 GitHub Release（附 zip / tar.gz / manifest / SHA256SUMS / latest.json）
 3. `npm publish` → [`@topmindspace/topmind-skills`](https://www.npmjs.com/package/@topmindspace/topmind-skills)（需仓库 Secret `NPM_TOKEN`，Automation token）
-4. **只保留最近 2 个 GitHub Release**，更旧的 Release 与 tag 会被删除
+4. **只保留最近 2 个 GitHub Release**（删旧 Release 与附件）。git tag 保留。
 
 | 渠道 | 触发 | 自动 |
 |------|------|------|

package/README.en.md

@@ -67,7 +67,7 @@ topmind-skills/              # pack root = repo root
 ```text
 User experience:     capture-first
 Data organization:   category-first + topic-emerges
-Content truth:       topmind-workspace/categories-and-topics
+Content truth:       topmind-workspace
 Capability model:    action-first
 Save settings:       auto | confirm
 Safety model:        reversible by default

package/README.md

@@ -67,7 +67,7 @@ topmind-skills/              # pack 根 = 仓库根
 ```text
 User experience:     capture-first
 Data organization:   category-first + topic-emerges
-Content truth:       topmind-workspace/categories-and-topics
+Content truth:       topmind-workspace
 Capability model:    action-first
 Save settings:       auto | confirm
 Safety model:        reversible by default
@@ -163,7 +163,7 @@ degradation: ../shared/capability-degradation.md
 
 ## 安装目标
 
-已打包的 skill 目录（7 个核心 + 2 个可选连接器 + 可选记账）可以符号链接或复制到 Claude Code、Codex、OpenCode、Hermes 等。  
+已打包的 skill 目录（7 个核心 + 2 个可选连接器 + 可选记账 + 可选公众号 `topmind-wechat`）可以符号链接或复制到 Claude Code、Codex、OpenCode、Hermes 等。  
 优先使用 npm / pack-aware 安装器，保证 `shared/` 与 `topmind-pack.json` 完整 — 见 [`INSTALL.md`](./INSTALL.md)。
 
 ```bash

package/evals/evals.json

@@ -1,7 +1,7 @@
 {
   "skill_name": "topmind",
   "daily_entry": "topmind",
-  "version": "4.13.3",
+  "version": "4.13.4",
   "evals": [
     {
       "id": 1,
@@ -112,28 +112,28 @@
       "prompt": "跑一遍 topmind-loop，做整体巡检。",
       "primary_skill": "topmind-loop",
       "extra_skills": [],
-      "expected_output": "Routes through topmind-loop (independent skill, not a maintain sub-action), walks all topics/inbox/archive scopes, applies well-defined fixes via UTR commands, writes decisions to .loop/topics.md, and flags architecture drift signals."
+      "expected_output": "Routes through topmind-loop (independent skill, not a maintain sub-action), walks all topics/inbox/archive scopes, applies well-defined fixes via UTR commands, writes decisions to .topmind/loop/topics.md, and flags architecture drift signals."
     },
     {
       "id": 16,
       "prompt": "继续 loop，从上次断点恢复。",
       "primary_skill": "topmind-loop",
       "extra_skills": [],
-      "expected_output": "Routes through topmind-loop, reads cursor from .loop/{scope}.md state files, resumes walk at the cursor item, updates done/total and cursor after each item, and appends new decisions."
+      "expected_output": "Routes through topmind-loop, reads cursor from .topmind/loop/{scope}.md state files, resumes walk at the cursor item, updates done/total and cursor after each item, and appends new decisions."
     },
     {
       "id": 17,
       "prompt": "loop 一下 20-研究 大类下的所有专题。",
       "primary_skill": "topmind-loop",
       "extra_skills": [],
-      "expected_output": "Routes through topmind-loop with scoped walk (only 20-研究 category topics), writes scoped progress to .loop/topics.md, applies UTR fixes for well-defined issues, escalates ambiguous decisions to Errors/Blockers."
+      "expected_output": "Routes through topmind-loop with scoped walk (only 20-研究 category topics), writes scoped progress to .topmind/loop/topics.md, applies UTR fixes for well-defined issues, escalates ambiguous decisions to Errors/Blockers."
     },
     {
       "id": 18,
       "prompt": "[schema validation] Verify all SKILL.md files have required frontmatter fields: name, version, description, action_category, triggers. Verify action_category values are within the enum. Verify name matches directory name in kebab-case.",
       "primary_skill": "topmind-maintain",
       "extra_skills": [],
-      "expected_output": "Validates frontmatter schema across all SKILL.md files (10 total: 7 core + 2 connector + 1 optional ledger). Reports any missing required fields, invalid action_category values, or name/directory mismatches."
+      "expected_output": "Validates frontmatter schema across all SKILL.md files (11 total: 7 core + 2 connector + optional ledger + optional wechat). Reports any missing required fields, invalid action_category values, or name/directory mismatches."
     },
     {
       "id": 19,

package/install-targets/claude-code.json

@@ -30,7 +30,7 @@
     "skills",
     "mcp"
   ],
-  "content_truth": "topmind-workspace/categories-and-topics",
+  "content_truth": "topmind-workspace",
   "host_must_not": [
     "change-content-truth",
     "add-daily-entrypoints",

package/install-targets/codex.json

@@ -31,7 +31,7 @@
     "cli",
     "mcp"
   ],
-  "content_truth": "topmind-workspace/categories-and-topics",
+  "content_truth": "topmind-workspace",
   "host_must_not": [
     "change-content-truth",
     "add-daily-entrypoints",

package/install-targets/hermes.json

@@ -31,7 +31,7 @@
     "cli",
     "mcp"
   ],
-  "content_truth": "topmind-workspace/categories-and-topics",
+  "content_truth": "topmind-workspace",
   "host_must_not": [
     "change-content-truth",
     "add-daily-entrypoints",

package/install-targets/mimocode.json

@@ -29,7 +29,7 @@
   "capabilities": [
     "skills"
   ],
-  "content_truth": "topmind-workspace/categories-and-topics",
+  "content_truth": "topmind-workspace",
   "host_must_not": [
     "change-content-truth",
     "add-daily-entrypoints",
@@ -43,8 +43,8 @@
     "workspace_root": "host-provided"
   },
   "notes": [
-    "MiMoCode / MiMo Desktop: project root .mimocode/skills, global root ~/.config/mimocode/skills.",
-    "Brand root ~/.agents/skills may also be scanned — canonical install already covers it.",
+    "MiMoCode / MiMo Desktop: project root .mimocode/skills, global root <mimocode-global-skills>.",
+    "Brand root <agents-skills> may also be scanned — canonical install already covers it.",
     "Single daily entry topmind. v4.3 progressive disclosure."
   ],
   "install": {
@@ -54,6 +54,6 @@
     "matrix": "node bin/install-skills.mjs add topmindspace/topmind-skills -g -a mimocode",
     "from_git": "node bin/install-skills.mjs add topmindspace/topmind-skills -g -a mimocode",
     "from_release": "node bin/install-skills.mjs add release:latest -g -a mimocode",
-    "single_dir": "node bin/install-skills.mjs add topmindspace/topmind-skills --dest ~/.config/mimocode/skills"
+    "single_dir": "node bin/install-skills.mjs add topmindspace/topmind-skills --dest <mimocode-global-skills>"
   }
 }

package/install-targets/opencode.json

@@ -32,7 +32,7 @@
     "commands",
     "plugin"
   ],
-  "content_truth": "topmind-workspace/categories-and-topics",
+  "content_truth": "topmind-workspace",
   "host_must_not": [
     "change-content-truth",
     "add-daily-entrypoints",

package/install-targets/universal.json

@@ -29,7 +29,7 @@
   "capabilities": [
     "skills"
   ],
-  "content_truth": "topmind-workspace/categories-and-topics",
+  "content_truth": "topmind-workspace",
   "host_must_not": [
     "change-content-truth",
     "add-daily-entrypoints",
@@ -42,7 +42,7 @@
     "skills_root": ".agents/skills"
   },
   "notes": [
-    "Open standard root: project ./.agents/skills, global ~/.agents/skills.",
+    "Open standard root: project ./.agents/skills, global <agents-skills>.",
     "Agents whose project skillsDir is .agents/skills (Codex, Cursor, Gemini CLI, Warp, Zed, …) read this directly.",
     "This is the canonical pack body. Private hosts symlink into it."
   ],
@@ -53,6 +53,6 @@
     "matrix": "node bin/install-skills.mjs add topmindspace/topmind-skills -g -a universal",
     "from_git": "node bin/install-skills.mjs add topmindspace/topmind-skills -g -a universal",
     "from_release": "node bin/install-skills.mjs add release:latest -g -a universal",
-    "single_dir": "node bin/install-skills.mjs add topmindspace/topmind-skills --dest ~/.agents/skills"
+    "single_dir": "node bin/install-skills.mjs add topmindspace/topmind-skills --dest <agents-skills>"
   }
 }

package/integrations/codex/README.md

@@ -1,19 +1,19 @@
 # topmind · Codex Integration
 
-[简体中文](../../README.md) · [English](../../README.en.md) · [Skills install](../../skills/INSTALL.md)
+[简体中文](../../README.md) · [English](../../README.en.md) · [Skills install](../../INSTALL.md)
 
 Codex consumes topmind through the same portable skill pack as other hosts.
 
 ## Recommended route
 
-1. Install or symlink the skills declared in `skills/topmind-pack.json` (10 modules: 7 core + 2 connectors + 1 ledger).  
+1. Install or symlink the skills declared in `topmind-pack.json`.  
 2. **Expose only `topmind`** as the daily user-facing entry; sub-skills are internal.  
 3. Use `topmind-cli` or MCP when available; **UTR is optional** — otherwise host file tools with the same filesystem contract.  
-4. Keep all user data under `topmind-workspace/categories-and-topics`. Codex session state must not become topmind content truth.
+4. Keep all user data under `topmind-workspace`. Codex session state must not become topmind content truth.
 
 ## Portable contract
 
-- Content truth is `topmind-workspace/categories-and-topics`.  
+- Content truth is `topmind-workspace`.  
 - Daily entry is `topmind`; no parallel daily entrypoints.  
 - **Desktop is not required** for the skill pack.  
 - Naming and frontmatter follow `PROJECT-MODEL.md` / `TOOLS.md`.  

package/integrations/hermes/README.md

@@ -1,19 +1,19 @@
 # topmind · Hermes Integration
 
-[简体中文](../../README.md) · [English](../../README.en.md) · [Skills install](../../skills/INSTALL.md)
+[简体中文](../../README.md) · [English](../../README.en.md) · [Skills install](../../INSTALL.md)
 
 Hermes consumes topmind as a portable skill pack, not a second workspace model.
 
 ## Recommended route
 
-1. Install or symlink the skills declared in `skills/topmind-pack.json`.  
+1. Install or symlink the skills declared in `topmind-pack.json`.  
 2. **Expose only `topmind`** as the daily entry; sub-skills route internally.  
 3. **UTR is optional** — use CLI/MCP when available; otherwise host file tools with the same filesystem contract.  
-4. Keep user data under `topmind-workspace/categories-and-topics`. Hermes-specific state must not become topmind content truth.
+4. Keep user data under `topmind-workspace`. Hermes-specific state must not become topmind content truth.
 
 ## Portable contract
 
-- Content truth is `topmind-workspace/categories-and-topics`.  
+- Content truth is `topmind-workspace`.  
 - Daily entry is `topmind`; no parallel daily entrypoints.  
 - **Desktop is not required** for the skill pack.  
 - Naming and frontmatter follow `PROJECT-MODEL.md` / `TOOLS.md`.  

package/integrations/opencode/README.md

@@ -1,16 +1,16 @@
 # topmind · OpenCode Integration
 
-[简体中文](../../README.md) · [English](../../README.en.md) · [Skills install](../../skills/INSTALL.md)
+[简体中文](../../README.md) · [English](../../README.en.md) · [Skills install](../../INSTALL.md)
 
 topmind should not fork OpenCode by default. Prefer a thin adapter layer.
 
 ## Recommended route
 
-1. Install the portable skill pack (`skills/topmind-pack.json` / pack-aware installer).  
+1. Install the portable skill pack (`topmind-pack.json` / pack-aware installer).  
 2. **Expose only `topmind`** as the daily user-facing entry — sub-skills are internal.  
 3. Expose UTR through MCP when available; **UTR is optional** otherwise use host file tools.  
 4. Optional OpenCode commands for common capture and doctor flows (`commands/`).  
-5. Keep content truth in `topmind-workspace/categories-and-topics`. OpenCode state must not become topmind content truth.
+5. Keep content truth in `topmind-workspace`. OpenCode state must not become topmind content truth.
 
 ## Config shape
 

package/integrations/opencode/plugins/topmind-plugin.ts

@@ -9,7 +9,7 @@ export default async function topmindPlugin(app?: OpenCodeApp) {
 
   return {
     name: "topmind",
-    contentTruth: "topmind-workspace/categories-and-topics",
+    contentTruth: "topmind-workspace",
     dailyEntry: "topmind",
     writesContent: false,
     capabilities: ["skills", "mcp", "commands", "plugin"],

package/integrations/opencode/tests/opencode-config-shape.test.mjs

@@ -57,7 +57,9 @@ test("OpenCode adapter follows the portable skill-pack contract", async () => {
   assert.match(readme, /content truth/u);
   assert.match(readme, /Expose only `topmind`|only daily.*`topmind`/iu);
 
-  assert.match(pluginSource, /contentTruth:\s*"topmind-workspace\/categories-and-topics"/u);
+  const truth = String(pack.portable_contract.content_truth).replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+  assert.match(pluginSource, new RegExp(`contentTruth:\\s*"${truth}"`));
+  assert.doesNotMatch(pluginSource, /categories-and-topics/u);
   assert.match(pluginSource, /writesContent:\s*false/u);
   assert.doesNotMatch(readme + "\n" + pluginSource, /should fork OpenCode|write topmind content directly:\s*true|writesContent:\s*true/u);
 });

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@topmindspace/topmind-skills",
-  "version": "4.13.3",
+  "version": "4.13.4",
   "description": "topmind Agent Skills pack — portable markdown skills for capture, organize, write, memory, maintain, loop.",
   "type": "module",
   "license": "MIT",

package/shared/document-ingest.md

@@ -59,7 +59,7 @@ ingest_converter: anydoc@0.1.8 | markitdown@… | pptx-ooxml
 
 失败时原件导入 + 说明笔记；可选原件进 `99-归档/ingest-originals/`。
 
-相关：网页剪藏分层见 [`long-url-capture.md`](./long-url-capture.md)；能力勾选见 [`../../docs/capture-clip-matrix.md`](../../docs/capture-clip-matrix.md)。
+相关：网页剪藏分层见 [`long-url-capture.md`](./long-url-capture.md)；能力勾选见 [`capture-clip-matrix.md`](https://github.com/topmindspace/topmind/blob/main/docs/capture-clip-matrix.md)。
 
 ## 格式与本机工具
 

package/shared/host-loading.md

@@ -11,15 +11,15 @@
 | **2 Activation** | 匹配到任务 | 完整 `SKILL.md` body | < 5k tokens 推荐 |
 | **3 Resources** | 步骤需要时 | `shared/*.md` · `references/*` · 可选 scripts | 按需 |
 
-**禁止**：启动时把 10 份 SKILL 全文 + shared 全部塞进 system prompt。
+**禁止**：启动时把全部 SKILL 全文 + shared 全部塞进 system prompt。
 
 ## 组合式 Pack 如何被 Host 理解
 
-topmind 是 **multi-skill pack**（1 router + 6 action + 2 connector + optional ledger），不是单文件 skill。业界对齐：
+topmind 是 **multi-skill pack**（目录与 `topmind-pack.json` 的 `skills` 一一对应，含 router、action、connector、可选 ledger 与可选 wechat write 子技能），不是单文件 skill。业界对齐：
 
 | 约定 | topmind 做法 |
 |------|----------------|
-| Agent Skills 单 skill = `{name}/SKILL.md` | 10 个目录各有 `SKILL.md` |
+| Agent Skills 单 skill = `{name}/SKILL.md` | 11 个目录各有 `SKILL.md` |
 | 部分 host 对 **zip 根** 找 `SKILL.md` | Release zip 根额外含 **router `SKILL.md`**（= `topmind/SKILL.md`） |
 | Pack 索引 | `skills.md` / `topmind-pack.json`（人类 + 机器） |
 | 共享资源 | `shared/` 与 skill 目录**同级** |
@@ -42,7 +42,7 @@ topmind-skills-<ver>/          ← Release zip 顶层
 │   └── SKILL.md
 ├── topmind-capture/
 │   └── SKILL.md
-├── …（其余 8 个，含可选 weread / x / ledger）
+├── …（其余 9 个，含可选 topmind-weread / topmind-x / topmind-ledger / topmind-wechat）
 ├── shared/                 ← 与 skill 同级！相对链接 ../shared/ 才能解析
 │   ├── capability-degradation.md
 │   ├── project-model-brief.md
@@ -56,7 +56,7 @@ topmind-skills-<ver>/          ← Release zip 顶层
 
 ### 错误形状
 
-- 只拷了 10 个目录、**没有** `shared/` → 渐进链接 404  
+- 只拷了 skill 目录、**没有** `shared/` → 渐进链接 404  
 - 把 monorepo 整仓当 skill 根 → host 扫到无关目录  
 - 只装 `topmind` 却期望 weread/x 触发 → connector 未装  
 - Host 报 `SKILL.md not found`：请用 `topmind-skills-*.zip`（含根 `SKILL.md`），或装到 `{skillsRoot}/topmind/SKILL.md`  
@@ -76,7 +76,7 @@ topmind-skills-<ver>/          ← Release zip 顶层
 node bin/install-skills.mjs add topmindspace/topmind-skills -g
 
 # 或手动
-cp -R skills/shared ~/.agents/skills/shared
+cp -R shared ~/.agents/skills/shared
 ```
 
 ## Router vs 子 skill（Host 路由策略）
@@ -86,7 +86,7 @@ cp -R skills/shared ~/.agents/skills/shared
 | 单一清晰动词（记一下 / 写一稿 / doctor） | 对应 `topmind-*` 子 skill |
 | 多意图 / 模糊 / 只说「topmind」 | `topmind` router |
 | Host **只装了** topmind 一个 skill | router 内嵌 Action Map 完成全部语义 |
-| 10 个全装 | description 消歧；router **不抢** 已匹配的子 skill |
+| 子 skill 全部安装 | description 消歧；router **不抢** 已匹配的子 skill |
 
 实现提示：discovery 阶段只比 `description` 文本相似度 / 关键词；**不要**用 triggers 数组作为唯一标准（triggers 是 topmind 扩展，开放标准不要求）。
 

package/topmind/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind
-version: 4.13.3
+version: 4.13.4
 description: >-
   topmind 总入口与多意图路由（类别/专题/笔记/交付）。Use when 用户说 topmind、意图模糊、或需要
   收→整→写 分步。单意图明确时直接用 topmind-capture|organize|write|memory|maintain|loop|weread|x|ledger|wechat。

package/topmind-capture/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-capture
-version: 4.13.3
+version: 4.13.4
 description: >-
   把链接、摘录、随手记收进动态周期本、Inbox 或专题。Use when 记一下、收进、剪藏、保存链接、capture、save URL/idea。
   Do NOT use for 整理本周、出稿、写我的情况、doctor/loop、微信读书、X.

package/topmind-ledger/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-ledger
-version: 4.13.3
+version: 4.13.4
 description: >-
   通用记账到记忆平面（默认个人/自己账本，用户自建账本与分类）。Use when 记账、记一笔、花了、存入、查看账单、账户余额。
   Do NOT use for 记一下到动态、待办、微信读书、发推、Feishu/lark-cli.

package/topmind-loop/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-loop
-version: 4.13.3
+version: 4.13.4
 description: >-
   可中断的工作区语义巡检（走专题/Inbox，.loop 断点续跑）。Use when 跑一遍 loop、巡检、整体体检、复盘、
   继续 loop、audit。Do NOT use for 快速 doctor（→maintain）、单专题整理、捕获、写作、仅记忆.

package/topmind-maintain/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-maintain
-version: 4.13.3
+version: 4.13.4
 description: >-
   确定性体检/清理/结构修复/回执恢复。Use when 快速体检、诊断、doctor、清理、修复、workspace check。
   Do NOT use for 整体巡检 loop、内容整理、捕获、写作、记忆.

package/topmind-memory/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-memory
-version: 4.13.3
+version: 4.13.4
 description: >-
   更新「我的情况」或周期反思。Use when 记住这个、更新我的情况、加到专题记忆、沉淀结论。
   Do NOT use for 捕获、仅总结（→organize）、整理本周正文、出稿、开/并专题、doctor/loop.

package/topmind-organize/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-organize
-version: 4.13.3
+version: 4.13.4
 description: >-
   整理本周动态、专题内整理/研究/分析/路由 Inbox。Use when 整理本周、整理、分析、研究、对比、总结要点、organize、summarize。
   Do NOT use for 首次捕获、最终出稿、仅写我的情况、快速 doctor、全库 loop.

package/topmind-pack.json

@@ -2,10 +2,10 @@
   "schema_version": 1,
   "name": "topmind",
   "label": "topmind",
-  "description": "topmind skill pack: single daily entry topmind + progressive disclosure (shared/ + references/), Agent Skills open-standard aligned, capture-first template-driven categories, optional weread/x connectors and optional ledger (记账). File-system-first content truth; UTR optional.",
+  "description": "topmind skill pack: single daily entry topmind + progressive disclosure (shared/ + references/), Agent Skills open-standard aligned, capture-first template-driven categories, optional connectors topmind-weread and topmind-x, optional topmind-ledger (记账), optional write sub-skill topmind-wechat. None of the optional skills is an extra user concept. File-system-first content truth; UTR optional.",
   "daily_entry": "topmind",
-  "version": "4.13.3",
-  "updated": "2026-09-24",
+  "version": "4.13.4",
+  "updated": "2026-09-25",
   "license": "MIT",
   "locales": {
     "default": "zh-CN",
@@ -72,7 +72,7 @@
       ],
       "runtime": "pure-markdown",
       "optional_dependencies": {
-        "utr": "node >= 18, optional deterministic CLI/MCP commands",
+        "utr": "node >= 20.11, optional deterministic CLI/MCP commands",
         "desktop": "Electron, optional GUI client (topmind-desktop/)"
       }
     },
@@ -216,7 +216,7 @@
     "topic_emerges": true,
     "action_first": true,
     "single_user_entry": true,
-    "content_truth": "topmind-workspace/categories-and-topics",
+    "content_truth": "topmind-workspace",
     "inbox_role": "unassigned-capture",
     "topic_role": "assigned-work",
     "loose_note_role": "unfiled-but-categorized"
@@ -279,7 +279,7 @@
   "portable_contract": {
     "requires_desktop": false,
     "requires_utr": false,
-    "content_truth": "topmind-workspace/categories-and-topics",
+    "content_truth": "topmind-workspace",
     "host_may_provide": [
       "skills",
       "cli",
@@ -313,6 +313,14 @@
     {
       "id": "claude-code",
       "path": "install-targets/claude-code.json"
+    },
+    {
+      "id": "mimocode",
+      "path": "install-targets/mimocode.json"
+    },
+    {
+      "id": "universal",
+      "path": "install-targets/universal.json"
     }
   ],
   "utr": {

package/topmind-wechat/DESIGN.md

@@ -1,6 +1,6 @@
 # topmind-wechat · 技能侧设计摘要
 
-> 全案见 [`docs/wechat-studio-DESIGN.md`](../../../docs/wechat-studio-DESIGN.md)。本文只记技能包边界。
+> 技能边界以本文件、`SKILL.md` 和 `references/` 为准。不要另做一份用浏览器本地存储的公众号编辑器。
 
 ## 定位
 

package/topmind-wechat/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-wechat
-version: 4.13.3
+version: 4.13.4
 description: >-
   公众号文章全生命周期子技能（write 族）：交付包、审校改写、质量三关、状态同步、微信内联排版与发布清单。
   支持 forward（底稿→公众号）、reverse（选题原创→回推）、站外拉取（在线精选站→reverse+pending）三条路径；

package/topmind-wechat/references/workflow.md

@@ -91,12 +91,12 @@ python3 scripts/md2wechat.py --input <包>/公众号稿.md --out-dir <包> \
 3. 提醒：外链只能进「阅读原文」；图已 base64 内嵌，占位符按清单补传  
 4. 需要进交付层时：`save-output` 拷贝终稿到 `88-交付/`（`YYYY-MM-DD-描述.ext`）
 
-## Desktop / 移动伴面
-
-| 步骤 | Desktop「公众号创作」 | wechat-studio/index.html |
-|------|----------------------|---------------------------|
-| 建包/列表 | 扫工作区包根 | 本地草稿箱 |
-| 改稿 | `api.ws.save` 回执 | 本地编辑 |
-| 三关 | AI 味 + lint + 事实勾选 | 同规则 JS 子集 |
-| 排版 | 主题预览 + 调脚本导出 | 约束子集预览 + 复制 |
-| 状态 | 调 `sync-status` 或镜像 | 本地镜像（回写以 Desktop 为准） |
+## Desktop
+
+| 步骤 | Desktop「公众号创作」 |
+|------|----------------------|
+| 建包/列表 | 扫工作区包根 |
+| 改稿 | 写闸回执 |
+| 三关 | AI 味 + lint + 事实勾选 |
+| 排版 | 主题预览 + 调脚本导出 |
+| 状态 | 调 `sync-status` |

package/topmind-weread/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-weread
-version: 4.13.3
+version: 4.13.4
 description: >-
   同步微信读书划线/想法/统计到专题。Use when 微信读书、划线同步、weread、读书笔记。
   Do NOT use for 普通 URL 捕获、仅整理、出稿、X.

package/topmind-write/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-write
-version: 4.13.3
+version: 4.13.4
 description: >-
   写作/润色/出稿到专题或交付层（role:delivery）。Use when 写、起草、续写、润色、出稿、交付、write、draft、deliver。
   Do NOT use for 仅捕获、仅结构整理、只写记忆、doctor/loop、社交连接器.

package/topmind-x/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: topmind-x
-version: 4.13.3
+version: 4.13.4
 description: >-
   X/Twitter：归档/搜索（Bearer）、发帖（xurl）、Agent 可用官方 MCP。Use when 发推、推特、twitter、x.com。
   Do NOT use for 微信读书、非 X 捕获、doctor、长文交付.