install

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/install — Setup & Health Check

/install — 安装设置与健康检查

Verify your Claude Code Toolkit installation, diagnose issues, and get oriented.

验证你的Claude Code Toolkit安装情况、诊断问题并快速了解工具概况。

When to Use

使用场景

After cloning the repo and running
```
install.sh
```
When something seems broken (hooks not firing, missing commands)
First time using the toolkit — to see what's available
After a
```
git pull
```
to verify nothing broke

克隆仓库并运行
```
install.sh
```
之后
工具出现异常时（钩子未触发、命令缺失等）
首次使用工具时——了解可用功能
执行
```
git pull
```
之后——验证工具是否正常运行

Instructions

操作步骤

Phase 1: DIAGNOSE

阶段1：诊断

Goal: Run deterministic health checks and report results.

Step 1: Run install-doctor.py

bash

python3 ~/.claude/scripts/install-doctor.py check

If the script is not found at

scripts/install-doctor.py

, try

~/.claude/scripts/install-doctor.py

Step 2: Interpret results

Result	Action
All checks pass	Skip to Phase 3 (Inventory)
`~/.claude` missing	Guide user to run `install.sh` — go to Phase 2
Components missing	Guide user to run `install.sh` — go to Phase 2
Hooks not configured	Guide user to run `install.sh` — go to Phase 2
Broken symlinks	Symlink targets moved. Re-run `install.sh --symlink --force`
Python deps missing	Run `pip install -r requirements.txt` from the repo directory
Permissions wrong	Run `chmod 755` on affected files

Step 3: Display results clearly

Show the check output to the user with a clear pass/fail summary. Use the raw script output — do not paraphrase or reformat excessively.

Gate: Health check complete. If issues found, proceed to Phase 2. If clean, skip to Phase 3.

目标：运行确定性健康检查并报告结果。

步骤1：运行install-doctor.py

bash

python3 ~/.claude/scripts/install-doctor.py check

如果未在

scripts/install-doctor.py

找到该脚本，尝试

~/.claude/scripts/install-doctor.py

。

步骤2：解读结果

结果	操作
所有检查通过	跳至阶段3（清单查看）
`~/.claude` 目录缺失	引导用户运行 `install.sh` ——进入阶段2
组件缺失	引导用户运行 `install.sh` ——进入阶段2
钩子未配置	引导用户运行 `install.sh` ——进入阶段2
符号链接损坏	符号链接目标已移动。重新运行 `install.sh --symlink --force`
Python依赖缺失	在仓库目录下运行 `pip install -r requirements.txt`
权限错误	对受影响文件执行 `chmod 755`

步骤3：清晰展示结果

向用户展示检查输出，并提供清晰的通过/失败总结。直接使用脚本的原始输出——不要过度改写或重新格式化。

节点：健康检查完成。若发现问题，进入阶段2；若一切正常，跳至阶段3。

Phase 2: FIX (only if issues found)

阶段2：修复（仅当发现问题时）

Goal: Guide the user through fixing detected issues.

Important: This phase is interactive. Show the user what needs fixing and let them choose.

Step 1: Determine if install.sh needs to run

~/.claude

is missing or components are not installed, the user needs to run install.sh. Tell them:

The toolkit hasn't been installed yet. Run this from the repo directory:

  ./install.sh --symlink     # recommended: updates with git pull
  ./install.sh --dry-run     # preview first

Wait for the user to confirm they've run it, then re-run the health check.

Step 2: Fix individual issues

For fixable issues (permissions, missing deps), offer to fix them:

bash

undefined

目标：引导用户修复检测到的问题。

注意：此阶段为交互式操作。向用户展示需要修复的内容并让他们选择操作。

步骤1：判断是否需要运行install.sh

如果

~/.claude

目录缺失或组件未安装，用户需要运行install.sh。告知他们：

工具尚未完成安装。请在仓库目录下运行以下命令：

  ./install.sh --symlink     # 推荐：配合git pull更新工具
  ./install.sh --dry-run     # 先预览安装内容

等待用户确认已运行命令后，重新执行健康检查。

步骤2：修复单个问题

对于可修复的问题（权限错误、依赖缺失），提供修复命令：

bash

undefined

Fix permissions

修复权限

find ~/.claude/hooks -name ".py" -exec chmod 755 {} ; find ~/.claude/scripts -name ".py" -exec chmod 755 {} ;

Install Python deps (from repo directory)

安装Python依赖（在仓库目录下执行）

pip install -r requirements.txt


Only run fixes the user approves.

**Step 3: Re-check**

After fixes, re-run:
```bash
python3 ~/.claude/scripts/install-doctor.py check

Gate: All checks pass. Proceed to Phase 3.

pip install -r requirements.txt


仅执行用户确认的修复操作。

**步骤3：重新检查**

修复完成后，重新运行：
```bash
python3 ~/.claude/scripts/install-doctor.py check

节点：所有检查通过。进入阶段3。

Phase 3: INVENTORY

阶段3：清单查看

Goal: Show the user what they have installed.

Step 1: Run inventory

bash

python3 ~/.claude/scripts/install-doctor.py inventory

Step 2: Display summary

Show the actual counts returned by

install-doctor.py inventory

— do not display hardcoded numbers. Present them as:

Your toolkit is ready. Here's what's installed:

  Agents:   [N] specialized domain experts
  Skills:   [N] workflow methodologies ([N] user-invocable)
  Hooks:    [N] automation hooks
  Commands: [N] slash commands
  Scripts:  [N] utility scripts

Gate: User sees their inventory. Proceed to Phase 3.5.

目标：向用户展示已安装的工具组件。

步骤1：运行清单命令

bash

python3 ~/.claude/scripts/install-doctor.py inventory

步骤2：展示汇总信息

显示

install-doctor.py inventory

返回的实际数量——不要使用硬编码数值。格式如下：

你的工具已准备就绪。已安装组件如下：

  Agents:   [N] 个专业领域专家Agent
  Skills:   [N] 个工作流方法（其中[N]个可由用户调用）
  Hooks:    [N] 个自动化钩子
  Commands: [N] 个斜杠命令
  Scripts:  [N] 个实用脚本

节点：用户查看完清单。进入阶段3.5。

Phase 3.5: MCP INVENTORY

阶段3.5：MCP清单查看

Goal: Show which MCP servers are available and their status.

Step 1: Run MCP registry check

bash

python3 ~/.claude/scripts/mcp-registry.py list

If the script is not found at

scripts/mcp-registry.py

, try

~/.claude/scripts/mcp-registry.py

Step 2: Display MCP status

Show the MCP inventory as:

MCP Servers:

  [✓] Chrome DevTools MCP  — Live browser debugging
      Paired skills: wordpress-live-validation
  [✓] Playwright MCP       — Automated browser testing
      Paired skills: wordpress-live-validation
  [✓] gopls MCP            — Go workspace intelligence
      Paired skills: go-code-review, go-testing, go-concurrency
  [✗] Context7 MCP         — Library documentation lookups
      Install: claude mcp add context7 -- npx @anthropic-ai/mcp-context7@latest

Use ✓ for connected MCPs and ✗ for missing ones. For missing MCPs, show the install command.

Gate: MCP inventory displayed. Proceed to Phase 4.

目标：展示可用的MCP服务器及其状态。

步骤1：运行MCP注册表检查

bash

python3 ~/.claude/scripts/mcp-registry.py list

如果未在

scripts/mcp-registry.py

找到该脚本，尝试

~/.claude/scripts/mcp-registry.py

。

步骤2：展示MCP状态

按以下格式展示MCP清单：

MCP服务器：

  [✓] Chrome DevTools MCP  — 实时浏览器调试
      关联Skill：wordpress-live-validation
  [✓] Playwright MCP       — 自动化浏览器测试
      关联Skill：wordpress-live-validation
  [✓] gopls MCP            — Go工作区智能分析
      关联Skill：go-code-review, go-testing, go-concurrency
  [✗] Context7 MCP         — 库文档查询
      安装命令：claude mcp add context7 -- npx @anthropic-ai/mcp-context7@latest

使用✓表示已连接的MCP，✗表示缺失的MCP。对于缺失的MCP，展示安装命令。

节点：MCP清单展示完成。进入阶段4。

Phase 4: ORIENT

阶段4：使用引导

Goal: Give the user their bearings — what to do first.

Step 1: Show the three essential commands

Getting started:

  /do [describe what you want]    — routes to the right agent + skill
  /comprehensive-review           — 20+ reviewer agents in 3 waves
  /install                        — run this again anytime to check health

Step 2: Show a few practical examples

Try these:

  /do debug this failing test
  /do review my Go code for quality
  /do write a blog post about [topic]
  /do create a voice profile from my writing samples

Step 3: Mention the docs

Documentation:
  docs/QUICKSTART.md   — 30-second overview
  docs/REFERENCE.md    — quick reference card

Gate: User is oriented. Installation complete.

目标：帮助用户快速上手——了解首次操作内容。

步骤1：展示三个核心命令

快速开始：

  /do [描述你的需求]    — 自动路由至对应的Agent + Skill
  /comprehensive-review           — 分3轮调用20+个评审Agent
  /install                        — 可随时重新运行以检查工具健康状态

步骤2：展示实用示例

尝试以下命令：

  /do debug this failing test
  /do review my Go code for quality
  /do write a blog post about [topic]
  /do create a voice profile from my writing samples

步骤3：提及文档位置

文档地址：
  docs/QUICKSTART.md   — 30秒快速概览
  docs/REFERENCE.md    — 速查手册

节点：用户完成引导。安装流程结束。

Error Handling

错误处理

Error: install-doctor.py not found

错误：install-doctor.py未找到

The script hasn't been installed yet. Check if the user is in the repo directory. If so, run it directly from

scripts/

. If not, guide them to clone and install first.

脚本尚未完成安装。检查用户是否处于仓库目录下。如果是，直接从

scripts/

目录运行脚本。如果不是，引导用户先克隆仓库并完成安装。

Error: Permission denied on install.sh

错误：install.sh权限不足

Run

chmod +x install.sh

first.

先运行

chmod +x install.sh

。

Error: Python not found

错误：Python未找到

The toolkit requires Python 3.10+. Guide the user to install Python for their platform.

工具要求Python 3.10及以上版本。引导用户为对应平台安装Python。