dsh-super-code 0.3.10 → 0.4.0

package/LICENSE

@@ -186,7 +186,7 @@
       same "printed page" as the copyright notice for easier
       identification within third-party archives.
 
-   Copyright [yyyy] [name of copyright owner]
+   Copyright 2026 ArtlexYoung
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

package/README.en.md

@@ -2,26 +2,30 @@
 
 [中文](README.md) · English
 
+[![npm total downloads](https://img.shields.io/npm/dt/dsh-super-code?label=npm%20downloads)](https://npmtrends.com/dsh-super-code) [![Latest npm release](https://img.shields.io/npm/v/dsh-super-code)](https://www.npmjs.com/package/dsh-super-code) [![License: Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-blue)](LICENSE)
+
 A smarter, leaner, faster, and easier-to-use coding plugin for DeepSeek Harness.
 
-On SWE-bench Pro hard-100, compared with the official Standard preset (same host DSH `0.2.1-rc.2`, same model, both on high reasoning effort and legacy budgets):
+**Evaluation results may vary slightly with model variability; refer to the latest tests.**
+
+On SWE-bench Pro hard-100, compared with the official Standard preset (same host DSH `0.2.1-rc.2`, same model, all on high reasoning effort):
 
-| | Standard + high | Super Code |
-|---|---:|---:|
-| Passed | 46/100 | **49/100** |
-| Net time per task | 20.9 min | **16.4 min** (-21.3%) |
-| Tokens per task | 7.81 M | **5.11 M** (-34.6%) |
-| Model calls per task | 92 | **69** (-25.5%) |
+| | Official Standard + high | Super Code 0.3.11 | Super Code 0.4.0 |
+|---|---:|---:|---:|
+| Passed | 47/100 | 50/100 (+3 tasks) | **52/100 (+5 tasks)** |
+| Time per task | 16.26 min | 16.69 min (+2.7%) | 17.66 min (+8.6%) |
+| Tokens per task | 7.718 M | 6.673 M (-13.5%) | 6.895 M (-10.7%) |
+| Model calls per task | 89.54 | 76.44 (-14.6%) | 79.42 (-11.3%) |
 
-Tested on version `0.3.5`. [Full results and limitations](docs/en/EvaluationResults.md)
+Tested on 2026-11-11; `0.4.0` is the frozen candidate used in this evaluation. All metrics use 100 tasks as the denominator and include original failed attempts and supplemental answers. Percentages are relative to Official Standard. Time includes tools and in-session waits, excluding grading; model calls count raw requests, including failed and auxiliary calls. Usage records are incomplete for **17, 10, and 13 tasks**, respectively; token values and percentage changes compare captured records only. [Historical results and methodology](docs/en/EvaluationResults.md)
 
 ## Install
 
-Supports Harness from `0.1.5-alpha.1`; see [compatibility](docs/en/Compatibility.md) for tested Web, official Desktop, and NEXT versions.
+Supports Harness `0.1.5-alpha.1` through `0.2.2-rc.2`, including `0.2.2-alpha.3`. Version 0.4.0 has been checked against all 27 Web releases in this range, official macOS Desktop `0.2.2-rc.2`, and NEXT `2.0.22-next` (bundled Harness `0.2.2-rc.1`); see [compatibility](docs/en/Compatibility.md) for tested versions and platform coverage.
 
 **Desktop**: open **Plugins → Add plugin** in the left sidebar, enter `dsh-super-code`, then click **Install** and **Enable now** (on older versions, reload or restart when prompted). Desktop includes its own runtime; Node.js is not required.
 
-**Web / CLI**: requires Node.js 22 or later.
+**Web / CLI**: the plugin requires Node.js 22 or later; Harness `0.2.2-rc.2` requires Node.js 24.9 or later.
 
 ```bash
 npm install -g @deepseek-ai/dsh            # skip if dsh is already installed
@@ -40,7 +44,7 @@ Desktop: enter the full path of the package in "Package name or address", then c
 Web: run the following command (use the actual file name):
 
 ```bash
-dsh plugin --profile web add ./dsh-super-code-0.3.10.tgz
+dsh plugin --profile web add ./dsh-super-code-0.4.0.tgz
 ```
 
 </details>
@@ -70,7 +74,7 @@ Select **Super Code** in the right sidebar. It contains three sections:
 - **Collaborative execution**: the Agent execution tree. Click a node to view details.
 - **Memory**: view, add, edit, and delete project memory and global preferences.
 
-The three sections open one at a time. Tasks opens by default; click the current heading to collapse it.
+Exactly one of the three sections stays open. Tasks opens by default; clicking its heading keeps it expanded.
 
 [![Task sidebar](docs/images/task-status-preview.png)](docs/images/task-status.png)
 
@@ -88,15 +92,22 @@ Screenshots use the current UI with sample data to illustrate the plugin.
 | Deletion protection | Directs Agents to use recoverable `trash` by default; can switch to strict confirmation |
 | Windows commands | Bundled Bash and common utilities, optionally applied to official presets |
 | MCP connection | Allows external MCP clients to delegate tasks to DSH. Off by default |
-| Frontend Skill | Loaded on demand for UI work to check alignment, hit areas, and browser behavior |
+| Preset Skills | Separate Skills and workflow groups for grill-me, OpenSpec, oil-ui, Matt Pocock Skills and PStack; enabled packages sync live to the DSH Skill directory, with a shortcut to open it |
 
 See the [usage guide](docs/en/UsageGuide.md).
 
+Bundled and adapted Skill sources are published under [builtin/skills](builtin/skills), including provenance, licenses, and companion resources. See the [layout and maintenance guide](builtin/README.md).
+
 ## Recent releases
 
+- **0.4.0**
+
+  1. Simplified task and memory tool calls with action-specific help and examples, plus direct reads of a named project-memory topic.
+  2. Added five opt-in Skills and workflow packages: grill-me, OpenSpec, oil-ui, Matt Pocock Skills, and PStack. Enabled packages sync to the DSH Skill directory, take effect in existing conversations, and can be found by package name in the native command menu.
+  3. Improved Settings and the task sidebar, expanded compatibility with newer Harness versions, and fixed numerous issues in preset recovery, MCP, deletion protection, and other features.
+
+- **0.3.11**: Fixed numerous bugs.
 - **0.3.10**: Added bundled Bash on Windows and an official-preset switch, with legacy-host compatibility and tooltip fixes.
-- **0.3.9**: Polished Settings and the task sidebar, improved MCP conversation control, fixed PTC child-tool approvals, and verified rc.4 Web compatibility.
-- **0.3.8**: Added MCP connection; external clients can delegate, inspect, continue, and cancel DSH conversations. Added a Home overview in Settings.
 
 [Full changelog](docs/en/Changelog.md)
 
@@ -106,7 +117,7 @@ Versions follow `major.performance.maintenance`: a change in the second number i
 
 ### Plugin comparison
 
-Full SWE-bench Pro hard-100 results, sorted by tasks passed. All runs use Harness `0.2.1-rc.2`, `deepseek-v4.1-flash-1017`, high reasoning effort, and legacy budgets. Each group reuses its environment continuously across tasks.
+Full SWE-bench Pro hard-100 results, sorted by tasks passed. All runs use Harness `0.2.1-rc.2`, `Evaluation model B`, high reasoning effort, and legacy budgets. Each group reuses its environment continuously across tasks.
 
 | Configuration | Passed/valid | Mean total tokens (M) | Mean time (min) | Mean model calls | Mean tool calls | Cache rate |
 |---|---:|---:|---:|---:|---:|---:|
@@ -127,8 +138,8 @@ M means one million tokens, including reasoning tokens. Time includes some in-ru
 
 ### Historical process
 
-1. **Old-host baseline** (DSH `0.1.5-rc.2`, `deepseek-v4.1-flash`, high): the Super Code `0.3.0` candidate passed 56 tasks, the official Standard preset 51, and Minimal 46. Relative to Minimal, time fell by 47.6% and tokens by 44.9%.
-2. **Regression after the host upgrade**: after upgrading to DSH `0.2.1-rc.2` and `deepseek-v4.1-flash-1017`, under default budgets Standard dropped to 46 and Super Code `0.3.4` to 41, below Standard. Investigation found the new host had tightened context budgets: the compression threshold fell from 800,000 to 550,464 tokens, and retained verbatim context from 160,000 to 98,560 tokens.
+1. **Old-host baseline** (DSH `0.1.5-rc.2`, `Evaluation model B`, high): the Super Code `0.3.0` candidate passed 56 tasks, the official Standard preset 51, and Minimal 46. Relative to Minimal, time fell by 47.6% and tokens by 44.9%.
+2. **Regression after the host upgrade**: after upgrading to DSH `0.2.1-rc.2` and `Evaluation model B`, under default budgets Standard dropped to 46 and Super Code `0.3.4` to 41, below Standard. Investigation found the new host had tightened context budgets: the compression threshold fell from 800,000 to 550,464 tokens, and retained verbatim context from 160,000 to 98,560 tokens.
 3. **Retest with legacy budgets**: both sides used legacy budgets. Standard stayed at 46 with 14.2% more tokens; Super Code `0.3.5` rose to 49 with only 0.4% more tokens and 19.4% less net time. Based on this, the plugin uses legacy budgets by default from `0.3.6`.
 4. **Cross-model check** (`glm-5.3-flash`, max): Super Code `0.3.4` passed 45 tasks and native ZCode 29. Super Code used less net time per task (32.9 vs 52.3 min) and more tokens (5.50 M vs 3.61 M).
 
@@ -153,6 +164,24 @@ These are earlier evaluation runs and are not pooled with the plugin comparison
 
 [Results and methodology](docs/en/EvaluationResults.md) · [All result tables](eval/results/swebench-pro-final-100/ResultsTables.md)
 
+## Star history
+
+<a href="https://star-history.com/#ArtlexYoung/dsh-super-code&amp;Date">
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="docs/images/star-history-dark.svg">
+    <img alt="dsh-super-code Star history" src="docs/images/star-history.svg" width="800">
+  </picture>
+</a>
+
+## npm download trends
+
+<a href="https://npmtrends.com/dsh-super-code">
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="docs/images/npm-downloads-dark.svg">
+    <img alt="dsh-super-code npm downloads over the last 13 weeks" src="docs/images/npm-downloads.svg" width="800">
+  </picture>
+</a>
+
 ## Acknowledgements
 
 This project recognizes the [LINUX DO](https://linux.do) community.
@@ -162,7 +191,10 @@ This project recognizes the [LINUX DO](https://linux.do) community.
 | [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) (MIT) | Host runtime: Agents, sessions, tools, PTC, Skills, UI extensions |
 | [ZCode](https://github.com/zai-org/ZCode) (Apache-2.0) | UI design |
 | [Agent Skills](https://agentskills.io/specification) | File format of the frontend Skill |
-| [Apple HIG](https://developer.apple.com/design/human-interface-guidelines), [WCAG 2.2](https://www.w3.org/TR/WCAG22/) | Interaction and accessibility principles in the frontend Skill |
+| [oil-ui](https://github.com/oil-oil/oil-ui) (MIT) | Bundled frontend design Skill: design workflow, style comparisons, and rendered UI review |
+| [OpenSpec](https://github.com/Fission-AI/OpenSpec) (MIT) | Complete specification-driven workflow Skill package |
+| [Matt Pocock Skills](https://github.com/mattpocock/skills) (MIT) | Complete engineering workflow Skill package |
+| [PStack](https://github.com/cursor/plugins/tree/main/pstack) (MIT) | Collaborative architecture, development, verification and review Skills |
 | [SWE-bench](https://github.com/SWE-bench/SWE-bench), [SWE-bench Pro](https://github.com/scaleapi/SWE-bench_Pro-os) (code MIT) | Tuning and evaluation tasks |
 | [js-yaml](https://github.com/nodeca/js-yaml), [Zod](https://github.com/colinhacks/zod) (MIT) | YAML parsing, data validation |
 

package/README.md

@@ -2,26 +2,30 @@
 
 中文 · [English](README.en.md)
 
+[![npm 总下载量](https://img.shields.io/npm/dt/dsh-super-code?label=npm%20downloads)](https://npmtrends.com/dsh-super-code) [![npm 当前发布版本](https://img.shields.io/npm/v/dsh-super-code)](https://www.npmjs.com/package/dsh-super-code) [![许可证：Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-blue)](LICENSE)
+
 一个更聪明、更节省、更快速、更好用的 DeepSeek Harness 编码插件。
 
-在 SWE-bench Pro hard-100 上，与官方标准预设对比（同一宿主 DSH `0.2.1-rc.2`、同一模型、均为 high 思考等级与旧预算）：
+**因大模型波动，不同评测结果会有略有变化，以最新测试为准。**
+
+在 SWE-bench Pro hard-100 上，与官方标准预设对比（同一宿主 DSH `0.2.1-rc.2`、同一模型、均为 high 思考等级）：
 
-| | 标准 + high | Super Code |
-|---|---:|---:|
-| 通过 | 46/100 | **49/100** |
-| 每题净耗时 | 20.9 分钟 | **16.4 分钟**（-21.3%） |
-| 每题 token | 781 万 | **511 万**（-34.6%） |
-| 每题模型调用 | 92 次 | **69 次**（-25.5%） |
+| | 官方标准 + high | Super Code 0.3.11 | Super Code 0.4.0 |
+|---|---:|---:|---:|
+| 通过 | 47/100 | 50/100（+3题） | **52/100（+5题）** |
+| 每题耗时 | 16.26 分钟 | 16.69 分钟（+2.7%） | 17.66 分钟（+8.6%） |
+| 每题 Token | 771.8 万 | 667.3 万（-13.5%） | 689.5 万（-10.7%） |
+| 每题模型调用 | 89.54 次 | 76.44 次（-14.6%） | 79.42 次（-11.3%） |
 
-基于版本 `0.3.5` 测试。[完整结果与限制](docs/EvaluationResults.md)
+2026-11-11 测试，`0.4.0` 为本次冻结候选。全部按100题汇总，含原失败尝试及补答，百分比相对官方标准。耗时含工具与会话等待，不含评分；模型调用按原始请求计，含失败及辅助调用。用量记录不完整题数依次为 **17、10、13题**，Token 数值及变化比例仅比较已采集记录。[历史评测结果与口径](docs/EvaluationResults.md)
 
 ## 安装
 
-支持 Harness `0.1.5-alpha.1` 起的版本；Web、官方桌面版和 NEXT 的实测范围见[兼容性记录](docs/Compatibility.md)。
+支持 Harness `0.1.5-alpha.1` ～ `0.2.2-rc.2`，包括 `0.2.2-alpha.3`。0.4.0 已逐版验证 27 个 Web 版本，并验证官方 macOS 桌面版 `0.2.2-rc.2`、NEXT `2.0.22-next`（内置 Harness `0.2.2-rc.1`）；各版本与平台的实测范围见[兼容性记录](docs/Compatibility.md)。
 
 **桌面版**：在左侧打开「插件 → 添加插件」，输入 `dsh-super-code`，依次点击「安装」和「立即启用」（旧版本请按提示重载或重启）。桌面版内置运行环境，无需安装 Node.js。
 
-**Web / 命令行版**：需要 Node.js 22 及以上版本。
+**Web / 命令行版**：插件要求 Node.js 22 及以上；Harness `0.2.2-rc.2` 要求 Node.js 24.9 及以上。
 
 ```bash
 npm install -g @deepseek-ai/dsh            # 已安装 dsh 命令可跳过
@@ -40,7 +44,7 @@ dsh plugin --profile web add dsh-super-code
 Web 版：运行以下命令（文件名以实际下载的为准）：
 
 ```bash
-dsh plugin --profile web add ./dsh-super-code-0.3.10.tgz
+dsh plugin --profile web add ./dsh-super-code-0.4.0.tgz
 ```
 
 </details>
@@ -70,7 +74,7 @@ dsh plugin --profile web add ./dsh-super-code-0.3.10.tgz
 - **协作执行**：Agent 执行树，点击节点可查看详情。
 - **记忆**：查看、新增、编辑和删除项目记忆与全局偏好。
 
-三个部分互斥展开，默认显示任务；点击当前标题可收起。
+三个部分始终只展开一项，默认显示任务；再次点击当前标题会保持展开。
 
 [![任务侧边栏](docs/images/task-status-preview.png)](docs/images/task-status.png)
 
@@ -88,15 +92,22 @@ dsh plugin --profile web add ./dsh-super-code-0.3.10.tgz
 | 删除保护 | 默认引导 Agent 使用可恢复的 `trash`，可切换为严格确认 |
 | Windows 命令环境 | 内置 Bash 与常用命令，可选应用到官方预设 |
 | MCP 连接 | 允许外部 MCP 客户端将任务委派给 DSH 执行，默认关闭 |
-| 前端 Skill | 开发页面时按需加载，检查对齐、点击区域和浏览器表现 |
+| 预设 Skill | grill-me、OpenSpec、oil-ui 与 Matt Pocock Skills、PStack 分组控制，开启后同步到 DSH Skill 目录、实时生效，支持快捷打开目录 |
 
 详见[使用说明](docs/UsageGuide.md)。
 
+内置及适配后的 Skill 源码统一公开在 [builtin/skills](builtin/skills)，保留各自的来源、许可证和配套资源；[目录与维护说明](builtin/README.md)。
+
 ## 最近更新
 
+- **0.4.0**
+
+  1. 简化任务与记忆工具的调用，提供按操作的帮助与示例，支持直接读取指定主题的项目记忆。
+  2. 新增 grill-me、OpenSpec、oil-ui、Matt Pocock Skills 和 PStack 五项预置 Skill 与工作流，默认关闭。启用后同步到 DSH Skill 目录，在已有对话中实时生效，支持按包名搜索原生命令。
+  3. 优化设置与任务侧栏，完善新版 Harness 兼容，修复预设恢复、MCP、删除保护等功能中的大量问题。
+
+- **0.3.11**：修复了大量 bug。
 - **0.3.10**：新增 Windows 内置 Bash 和官方预设开关，修复旧宿主兼容与提示气泡样式。
-- **0.3.9**：美化设置和任务侧栏，完善 MCP 对话控制并修复 PTC 子工具审批，验证 rc.4 Web 兼容性。
-- **0.3.8**：新增 MCP 连接，外部客户端可委派、查看、继续和取消 DSH 对话；设置页新增首页概览。
 
 [完整版本记录](docs/Changelog.md)
 
@@ -106,7 +117,7 @@ dsh plugin --profile web add ./dsh-super-code-0.3.10.tgz
 
 ### 插件方案对比
 
-SWE-bench Pro hard-100 完整成绩，按通过数量降序排列。统一使用 Harness `0.2.1-rc.2`、`deepseek-v4.1-flash-1017`、high、旧预算，各组内部连续复用环境。
+SWE-bench Pro hard-100 完整成绩，按通过数量降序排列。统一使用 Harness `0.2.1-rc.2`、`评测模型 B`、high、旧预算，各组内部连续复用环境。
 
 | 方案 | 通过/有效 | 平均总 Token（M） | 平均耗时（分钟） | 平均模型调用 | 平均工具调用 | 缓存率 |
 |---|---:|---:|---:|---:|---:|---:|
@@ -127,8 +138,8 @@ M 为百万 Token，包含思考 Token；耗时包含部分运行内服务等待
 
 ### 历史评测过程
 
-1. **旧宿主基线**（DSH `0.1.5-rc.2`，`deepseek-v4.1-flash`，high）：Super Code `0.3.0` 候选通过 56 题，官方标准预设 51 题，minimal 46 题。相对 minimal，耗时减少 47.6%，token 减少 44.9%。
-2. **宿主升级后效果下降**：升级到 DSH `0.2.1-rc.2` 和 `deepseek-v4.1-flash-1017` 后，在默认预算下标准降至 46 题，Super Code `0.3.4` 降至 41 题，低于标准。排查发现新宿主收紧了上下文预算：压缩阈值由 800,000 降至 550,464 token，保留原文由 160,000 降至 98,560 token。
+1. **旧宿主基线**（DSH `0.1.5-rc.2`，`评测模型 B`，high）：Super Code `0.3.0` 候选通过 56 题，官方标准预设 51 题，minimal 46 题。相对 minimal，耗时减少 47.6%，token 减少 44.9%。
+2. **宿主升级后效果下降**：升级到 DSH `0.2.1-rc.2` 和 `评测模型 B` 后，在默认预算下标准降至 46 题，Super Code `0.3.4` 降至 41 题，低于标准。排查发现新宿主收紧了上下文预算：压缩阈值由 800,000 降至 550,464 token，保留原文由 160,000 降至 98,560 token。
 3. **恢复旧预算后复测**：双方均使用旧预算。标准仍为 46 题，token 增加 14.2%；Super Code `0.3.5` 提升至 49 题，token 仅增加 0.4%，净耗时减少 19.4%。据此，`0.3.6` 起插件默认使用旧预算。
 4. **跨模型验证**（`glm-5.3-flash`，max）：Super Code `0.3.4` 通过 45 题，原生 ZCode 通过 29 题；Super Code 每题净耗时更少（32.9 vs 52.3 分钟），token 更多（550 万 vs 361 万）。
 
@@ -153,6 +164,24 @@ M 为百万 Token，包含思考 Token；耗时包含部分运行内服务等待
 
 [评测结果与口径](docs/EvaluationResults.md) · [完整成绩表](eval/results/swebench-pro-final-100/ResultsTables.md)
 
+## Star 趋势
+
+<a href="https://star-history.com/#ArtlexYoung/dsh-super-code&amp;Date">
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="docs/images/star-history-dark.svg">
+    <img alt="dsh-super-code Star history" src="docs/images/star-history.svg" width="800">
+  </picture>
+</a>
+
+## npm 下载趋势
+
+<a href="https://npmtrends.com/dsh-super-code">
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="docs/images/npm-downloads-dark.svg">
+    <img alt="dsh-super-code 最近 13 周 npm 下载趋势" src="docs/images/npm-downloads.svg" width="800">
+  </picture>
+</a>
+
 ## 致谢
 
 本项目认可 [LINUX DO](https://linux.do) 社区。
@@ -162,7 +191,10 @@ M 为百万 Token，包含思考 Token；耗时包含部分运行内服务等待
 | [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness)（MIT） | 运行宿主：Agent、会话、工具、PTC、Skill、界面扩展 |
 | [ZCode](https://github.com/zai-org/ZCode)（Apache-2.0） | 界面设计 |
 | [Agent Skills](https://agentskills.io/specification) | 前端 Skill 的文件格式 |
-| [Apple HIG](https://developer.apple.com/design/human-interface-guidelines)、[WCAG 2.2](https://www.w3.org/TR/WCAG22/) | 前端 Skill 的交互与无障碍原则 |
+| [oil-ui](https://github.com/oil-oil/oil-ui)（MIT） | 内置前端设计 Skill，提供设计流程、风格对比与画面评审 |
+| [OpenSpec](https://github.com/Fission-AI/OpenSpec)（MIT） | 完整的规范驱动工作流 Skill 包 |
+| [Matt Pocock Skills](https://github.com/mattpocock/skills)（MIT） | 完整的工程工作流 Skill 包 |
+| [PStack](https://github.com/cursor/plugins/tree/main/pstack)（MIT） | 架构、开发、验证与评审的协作 Skill 包 |
 | [SWE-bench](https://github.com/SWE-bench/SWE-bench)、[SWE-bench Pro](https://github.com/scaleapi/SWE-bench_Pro-os)（代码 MIT） | 调优题与评测题 |
 | [js-yaml](https://github.com/nodeca/js-yaml)、[Zod](https://github.com/colinhacks/zod)（MIT） | YAML 解析、数据校验 |
 

package/builtin/README.md

@@ -0,0 +1,22 @@
+# 内置 Skill
+
+本项目内置及面向 DSH 适配的 Skill 源码统一放在 [`skills/`](skills/)，正文、参考资料、脚本和许可证随仓库公开。
+
+| 包 | 可用内容 | 来源与许可证 |
+|---|---|---|
+| oil-ui | [设计流程与配套资源](skills/oil-ui/SKILL.md) | [来源与改动](skills/oil-ui/UPSTREAM.md) · [MIT](skills/oil-ui/LICENSE) |
+| OpenSpec | [规范工作流](skills/openspec/skills/) | [来源](skills/openspec/UPSTREAM.md) · [MIT](skills/openspec/LICENSE) |
+| Matt Pocock Skills | [工程工作流](skills/mattpocock/skills/) | [来源](skills/mattpocock/UPSTREAM.md) · [MIT](skills/mattpocock/LICENSE) |
+| PStack | [DSH 适配版](skills/pstack/dsh/skills/)及[宿主操作说明](skills/pstack/dsh/DSH.md) | [来源与适配](skills/pstack/UPSTREAM.md) · [MIT](skills/pstack/LICENSE) |
+
+设置按 Skill 和工作流分组：Skill 为 grill-me、OpenSpec、oil-ui，工作流为 Matt Pocock Skills、PStack。grill-me 独立开关复用 Matt 包中的 [grill-me](skills/mattpocock/skills/grill-me/SKILL.md) 与 [grilling](skills/mattpocock/skills/grilling/SKILL.md)，完整 Matt 包也包含这两项。开启后，完整资源同步到 DSH 用户 Skill 目录的 `dsh-super-code/` 下，保留包内路径关系，并由原生加载器实时提供启用项。开关不改动项目；项目和用户自己的 Skill 仍优先加载。旧用户预设保留原有资源，用于旧插件和独立加载；当前插件统一管理 DSH 目录中的副本。
+
+PStack 的固定上游原件保存在 `skills/pstack/`，可运行的 DSH 适配版位于 `skills/pstack/dsh/`。薄适配器仍在 [`runtime/skills/pstack/`](../runtime/skills/pstack/README.md)：修改适配规则或覆盖文件后运行 `npm run skills:build`，通过 `npm run skills:check` 核对生成内容、资源引用及来源哈希。升级上游时先审查新版本，不直接修改固定原件进行适配。
+
+npm 包包含可运行的内置 Skill、配套资源与许可证；PStack 的重复上游原件和构建期适配器保留在 Git，不重复装入 npm。各第三方内容遵循对应目录中的许可证，项目的适配代码按本仓库许可证公开。
+
+## Bundled Skill sources
+
+Bundled and DSH-adapted Skills are published under [`skills/`](skills/), with their instructions, resources, helpers, provenance, and licenses. The table above links to each package's usable entrypoints and source notes.
+
+PStack's pinned originals remain in `skills/pstack/`; its runnable DSH edition lives in `skills/pstack/dsh/`. The build-time adapter stays in [`runtime/skills/pstack/`](../runtime/skills/pstack/README.md). Run `npm run skills:build` after adapter changes and `npm run skills:check` to verify generated files and resource references. npm ships the runnable edition; the upstream originals and build-time adapter stay in Git. Existing switches, invocation rules, project/user precedence, and legacy preset layouts are preserved.

package/builtin/skills/mattpocock/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Matt Pocock
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/builtin/skills/mattpocock/UPSTREAM.md

@@ -0,0 +1,11 @@
+# Matt Pocock Skills
+
+Source: https://github.com/mattpocock/skills
+Revision: `4ae2317ddb22b046e26455e4f9c9b3aaaf62212d`
+Official plugin version: **1.3.2**. License: [MIT](LICENSE).
+
+Includes the complete 27-Skill distribution from `.claude-plugin/plugin.json`, with all references, templates and agent metadata. The upstream category directories are flattened into `skills/` for the native DSH filesystem provider. Skill contents and invocation flags are unchanged. Unreleased `in-progress`, `misc` and repository development instructions are outside the official plugin distribution.
+
+One plugin setting enables the entire package. No upstream installer, setup workflow or helper script runs on startup or when toggled. Skills load on demand and remain subject to the user's instructions, available host tools and DSH permissions. GitHub/GitLab workflows require the corresponding authenticated CLI and authorization for their writes; setup is an explicit user workflow. The wizard template may write `.env` and CI secrets when deliberately generated and run; it is never executed by the plugin. Updates ship with dsh-super-code, not an independent updater.
+
+See `distribution.json` for the exact source-to-bundle mapping and SHA-256 hashes.