openspec-playwright 0.3.60 → 0.3.62
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +0 -14
- package/README.zh-CN.md +0 -14
- package/employee-standards.md +13 -19
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -121,14 +121,6 @@ openspec-pw uninstall # Remove integration from the project
|
|
|
121
121
|
- **Cline**: merged into `.cline/mcp.json` under `mcpServers.playwright = { "command": "npx", "args": ["@playwright/mcp@latest"] }`
|
|
122
122
|
- **Cursor**: merged into `.cursor/mcp.json` under `mcpServers.playwright = { "command": "npx", "args": ["@playwright/mcp@latest"] }`
|
|
123
123
|
|
|
124
|
-
**Optional** — enhance your AI coding assistant with Superpowers methodology:
|
|
125
|
-
|
|
126
|
-
- **Superpowers**: a complete development methodology plugin for Claude Code. It enhances the OpenSpec workflow (propose → apply → verify) with conversational spec exploration, TDD discipline, and subagent-driven implementation. Superpowers does **not** replace OpenSpec — the E2E verification pipeline (`/opsx:e2e`, `openspec-pw doctor/explore/run`) remains unchanged.
|
|
127
|
-
```bash
|
|
128
|
-
/plugin install superpowers@claude-plugins-official
|
|
129
|
-
```
|
|
130
|
-
See [github.com/obra/superpowers](https://github.com/obra/superpowers) for details.
|
|
131
|
-
|
|
132
124
|
Browser exploration is provided out of the box by Playwright MCP and `openspec-pw explore`; no extra browser tool is needed.
|
|
133
125
|
|
|
134
126
|
## What `openspec-pw init` Does
|
|
@@ -175,12 +167,6 @@ Run through these steps in order when using the E2E workflow for the first time:
|
|
|
175
167
|
|
|
176
168
|
Run with `--json` for machine-readable output.
|
|
177
169
|
|
|
178
|
-
**Optional — enhance your AI coding assistant with Superpowers methodology:**
|
|
179
|
-
|
|
180
|
-
| Step | Command | If it fails |
|
|
181
|
-
|------|---------|-------------|
|
|
182
|
-
| A. Install Superpowers | `/plugin install superpowers@claude-plugins-official` | See [github.com/obra/superpowers](https://github.com/obra/superpowers) for alternative install methods |
|
|
183
|
-
|
|
184
170
|
## App Server Detection
|
|
185
171
|
|
|
186
172
|
Generated `playwright.config.ts` automatically detects the app URL in this priority order:
|
package/README.zh-CN.md
CHANGED
|
@@ -23,14 +23,6 @@ npm install -g openspec-playwright@latest
|
|
|
23
23
|
- **Cline**:合并到 `.cline/mcp.json` 的 `mcpServers.playwright = { "command": "npx", "args": ["@playwright/mcp@latest"] }`
|
|
24
24
|
- **Cursor**:合并到 `.cursor/mcp.json` 的 `mcpServers.playwright = { "command": "npx", "args": ["@playwright/mcp@latest"] }`
|
|
25
25
|
|
|
26
|
-
**可选** — 用 Superpowers 增强 AI 编码助手的能力:
|
|
27
|
-
|
|
28
|
-
- **Superpowers**:完整的 AI 编码助手开发方法论插件。它在 OpenSpec 工作流(propose → apply → verify)阶段增强 AI 能力:对话式 spec 探索、TDD 纪律、subagent 并行实现。Superpowers **不替换 OpenSpec**,E2E 验证管线(`/opsx:e2e`、`openspec-pw doctor/explore/run` 等)保持不变。
|
|
29
|
-
```bash
|
|
30
|
-
/plugin install superpowers@claude-plugins-official
|
|
31
|
-
```
|
|
32
|
-
详见 [github.com/obra/superpowers](https://github.com/obra/superpowers)
|
|
33
|
-
|
|
34
26
|
浏览器探索能力由 Playwright MCP 和 `openspec-pw explore` 内置提供,无需额外工具。
|
|
35
27
|
|
|
36
28
|
## 初始化
|
|
@@ -175,12 +167,6 @@ openspec-pw uninstall # 移除项目中的集成
|
|
|
175
167
|
|
|
176
168
|
加 `--json` 参数输出机器可读格式。
|
|
177
169
|
|
|
178
|
-
**可选 — 用 Superpowers 增强 AI 编码助手:**
|
|
179
|
-
|
|
180
|
-
| 步骤 | 命令 | 失败时快速修复 |
|
|
181
|
-
|------|------|----------------|
|
|
182
|
-
| A. 安装 Superpowers | `/plugin install superpowers@claude-plugins-official` | 详见 [github.com/obra/superpowers](https://github.com/obra/superpowers) |
|
|
183
|
-
|
|
184
170
|
## 认证配置
|
|
185
171
|
|
|
186
172
|
如果你的应用需要登录,配置一次凭证后,所有测试自动以已登录状态运行。
|
package/employee-standards.md
CHANGED
|
@@ -2,6 +2,12 @@
|
|
|
2
2
|
|
|
3
3
|
---
|
|
4
4
|
|
|
5
|
+
## CodeGraph 优先 🔴
|
|
6
|
+
|
|
7
|
+
存在 `.codegraph/` 时:结构/定位问题先 `codegraph_explore`,字面文本/已打开文件用 grep/read,自行判断。无 `.codegraph/` 跳过。
|
|
8
|
+
|
|
9
|
+
---
|
|
10
|
+
|
|
5
11
|
## 0. 适用范围
|
|
6
12
|
|
|
7
13
|
**约定**:动手前先读 `openspec/config.yaml`(技术栈、结构、约定、约束等),无内容则忽略。
|
|
@@ -42,7 +48,7 @@
|
|
|
42
48
|
|
|
43
49
|
**REQUIRE**
|
|
44
50
|
- linter/typechecker 不存在 → 告知用户并建议安装
|
|
45
|
-
- mock 数据 / fixture → 参见 §
|
|
51
|
+
- mock 数据 / fixture → 参见 §5 数据编撰禁令
|
|
46
52
|
- 涉及 API 定义 → 查阅真实 OpenAPI/MCP 定义并标注来源
|
|
47
53
|
|
|
48
54
|
---
|
|
@@ -63,24 +69,12 @@
|
|
|
63
69
|
|
|
64
70
|
---
|
|
65
71
|
|
|
66
|
-
## 3.
|
|
67
|
-
|
|
68
|
-
> **[项目特定]** 以下仅适用于 openspec-playwright 项目结构。WHY:这些约束一违反即产生 review 难发现的 bug。Anthropic 称其为最高杠杆的 agent 代码质量措施。
|
|
69
|
-
|
|
70
|
-
- 🔴 CLI 命令向 `src/index.ts` 注册,`src/commands/` 下按职责分文件
|
|
71
|
-
- 🔴 模板文件放 `templates/`,不与其他源码混放
|
|
72
|
-
- 🟡 OpenSpec change artifacts 放 `changes/<name>/`,遵循命名规范
|
|
73
|
-
- 🟡 E2E 测试生成代码放 `e2e/`(如 `e2e/auth.setup.ts`)
|
|
74
|
-
- 🔴 文档同步规则:改 `src/commands/*.ts`/`src/index.ts` 时必须同步更新 README + CHANGELOG
|
|
75
|
-
|
|
76
|
-
---
|
|
77
|
-
|
|
78
|
-
## 4. 工具使用
|
|
72
|
+
## 3. 工具使用
|
|
79
73
|
|
|
80
74
|
> WHY:工具使用方式直接影响搜索覆盖面和结果准确性。
|
|
81
75
|
|
|
82
76
|
**DO**
|
|
83
|
-
- 🟡
|
|
77
|
+
- 🟡 搜索分层:结构性问题用 CodeGraph(见文首);字面文本用全文搜索;文件名模式用文件名匹配。跳过依赖目录和缓存目录(调试依赖时除外),搜子目录时按需缩小
|
|
84
78
|
- 🟡 重命名覆盖:调用、类型、字符串、import、barrel file、测试 mock,不得假设一次覆盖
|
|
85
79
|
- 🟡 联网调研优先 agent-reach skill
|
|
86
80
|
- 🟡 涉及前端 UI 设计时,按序使用:frontend-design skill 定方向 → ui-ux-pro-max skill 选风格 → web-design-guidelines skill 审查,三步组合避免"千篇一律 AI 风"
|
|
@@ -98,7 +92,7 @@
|
|
|
98
92
|
|
|
99
93
|
---
|
|
100
94
|
|
|
101
|
-
##
|
|
95
|
+
## 4. 大规模任务
|
|
102
96
|
|
|
103
97
|
> WHY:200+ 行直接修改缺乏评审,OpenSpec 工作流强制提案→设计→评审→实现。
|
|
104
98
|
|
|
@@ -114,11 +108,11 @@
|
|
|
114
108
|
| E2E | `/opsx:e2e` / `/opsx-e2e` | 测试生成 + Healer 验证通过 |
|
|
115
109
|
| 归档 | `/opsx:archive` | specs 更新,归档完成 |
|
|
116
110
|
|
|
117
|
-
所有阶段由用户手动触发,**AI 不自动进入下一阶段**。
|
|
111
|
+
所有阶段由用户手动触发,**AI 不自动进入下一阶段**。
|
|
118
112
|
|
|
119
113
|
---
|
|
120
114
|
|
|
121
|
-
##
|
|
115
|
+
## 5. 数据编撰禁令
|
|
122
116
|
|
|
123
117
|
> WHY:AI 有"填空"倾向——缺乏数据时编造看似合理的值,引入静默 bug。
|
|
124
118
|
|
|
@@ -136,7 +130,7 @@
|
|
|
136
130
|
|
|
137
131
|
---
|
|
138
132
|
|
|
139
|
-
##
|
|
133
|
+
## 6. 临时文件管理
|
|
140
134
|
|
|
141
135
|
> WHY:临时文件散落在项目根目录而非集中存放,会在 commit 时被意外加入版本控制(导致仓库膨胀),或积累到难以清理(占用磁盘空间)。本节建立可自动化管理协议,避免这两类问题。
|
|
142
136
|
|