安装与启动

npm install -g @gitlawb/openclaude
openclaude

快速配置

OpenAI

export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=sk-your-key
export OPENAI_MODEL=gpt-4o

Ollama 本地

export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_BASE_URL=http://localhost:11434/v1
export OPENAI_MODEL=qwen2.5-coder:7b

DeepSeek

export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=sk-...
export OPENAI_BASE_URL=https://api.deepseek.com/v1
export OPENAI_MODEL=deepseek-chat

更多 Provider 示例

# LM Studio (本地)
export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_BASE_URL=http://localhost:1234/v1
export OPENAI_MODEL=your-model-id

# Together AI
export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=...
export OPENAI_BASE_URL=https://api.together.xyz/v1
export OPENAI_MODEL=meta-llama/Llama-3.3-70B-Instruct-Turbo

# Groq
export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=gsk_...
export OPENAI_BASE_URL=https://api.groq.com/openai/v1
export OPENAI_MODEL=llama-3.3-70b-versatile

# Mistral
export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=...
export OPENAI_BASE_URL=https://api.mistral.ai/v1
export OPENAI_MODEL=mistral-large-latest

# Azure OpenAI
export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=your-azure-key
export OPENAI_BASE_URL=https://your-resource.openai.azure.com/openai/deployments/your-deployment/v1
export OPENAI_MODEL=gpt-4o

# Codex via ChatGPT auth
export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_MODEL=codexplan  # 或 codexspark（更快）
export CODEX_API_KEY=...  # 可选，已有 ~/.codex/auth.json 则自动读取

slash 命令速查

核心命令

会话管理

Git 操作

工具与插件

配置与诊断

其他命令

配置目录

环境变量

Provider 选择

OpenAI 兼容模式

Anthropic

Google Gemini

GitHub Models

AWS Bedrock

Google Vertex AI

Codex 特定

Web 搜索

重试与超时

调试

其他

settings.json 配置

Agent 路由

{
  "agentModels": {
    "deepseek-chat": {
      "base_url": "https://api.deepseek.com/v1",
      "api_key": "sk-your-key"
    }
  },
  "agentRouting": {
    "Explore": "deepseek-chat",
    "Plan": "gpt-4o",
    "general-purpose": "gpt-4o",
    "default": "gpt-4o"
  }
}

权限配置

{
  "permissions": {
    "allow": [{ "command": "npm install" }],
    "deny": [{ "command": "rm -rf /" }],
    "defaultMode": "ask"
  }
}

Hooks 配置

{
  "hooks": {
    "BeforeThinking": [{ "command": "echo 'thinking...'" }],
    "AfterThinking": [{ "command": "echo 'done'" }]
  }
}

工作目录配置

{
  "additionalDirectories": ["/path/to/include"]
}

其他配置项

Skill 系统

安装 Skills

# GitHub 速记
npx skills add <owner/repo>

# 安装到全局
npx skills add vercel-labs/agent-skills -g

# 安装特定 skill
npx skills add vercel-labs/agent-skills --skill frontend-design

# 列出可用 skills
npx skills add vercel-labs/agent-skills --list

# 列出已安装
npx skills list

复用 Skills

skill: cheatsheet-maker
agentskill: icon-complete-skill

• skills/
• .claude/skills/
• .agents/skills/

• INSTALL_INTERNAL_SKILLS=1

Provider Profile 脚本

# 引导式初始化（优先本地 Ollama，否则 OpenAI）
bun run profile:init

# 推荐最佳 provider/model
bun run profile:recommend -- --goal coding --benchmark

# 自动应用最佳配置
bun run profile:auto -- --goal latency

# 启动已保存的 profile
bun run dev:profile

# Codex 引导
bun run profile:codex

# 快速 profile
bun run profile:fast
bun run profile:code

诊断与验证

bun run smoke                  # 快速启动检查
bun run doctor:runtime         # 验证 provider 环境
bun run doctor:runtime:json    # 机器可读诊断
bun run doctor:report          # 生成诊断报告
bun run hardening:check       # 本地硬化检查
bun run hardening:strict       # 严格硬化（含 typecheck）

gRPC 服务

npm run dev:grpc              # 启动 gRPC 服务器（默认 50051）
npm run dev:grpc:cli          # 运行 gRPC CLI 客户端

VS Code 扩展

开发命令

bun run dev              # 开发模式
bun test                 # 运行测试
bun run test:coverage     # 测试覆盖率
bun run build            # 构建
bun run verify:privacy   # 隐私验证

仓库结构

src/                     核心 CLI/运行时
scripts/                 构建、验证脚本
docs/                    设置与贡献文档
python/                  Python helpers 和测试
vscode-extension/        VS Code 扩展
bin/                     CLI 入口点