mcp-client

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Universal MCP Client

通用MCP Client

Connect to any MCP server with progressive disclosure - load tool schemas on-demand instead of dumping thousands of tokens into context upfront.

支持渐进式披露连接任意MCP服务器——按需加载工具 schema，无需预先向上下文注入数千个token。

Skill Location

Skill 位置

This skill is located at:

.claude/skills/mcp-client/

Script path:

.claude/skills/mcp-client/scripts/mcp_client.py

该Skill位于：

.claude/skills/mcp-client/

脚本路径：

.claude/skills/mcp-client/scripts/mcp_client.py

Configuration

配置

The script looks for config in this order:

```
MCP_CONFIG_PATH
```
env var (custom path)
references/mcp-config.json
(this skill's config - recommended)
```
.mcp.json
```
in project root
```
~/.claude.json
```

Your config file:

.claude/skills/mcp-client/references/mcp-config.json

Edit this file to add your API keys. The example file (

example-mcp-config.json

) is kept as a reference template.

If the user hasn't provided their Zapier API key yet, ask them for it.

脚本按以下顺序查找配置：

```
MCP_CONFIG_PATH
```
环境变量（自定义路径）
references/mcp-config.json
（本Skill的配置文件——推荐使用）
项目根目录下的
```
.mcp.json
```
```
~/.claude.json
```

你的配置文件：

.claude/skills/mcp-client/references/mcp-config.json

编辑此文件添加你的API密钥。示例文件（

example-mcp-config.json

）可作为参考模板。

如果用户尚未提供Zapier API密钥，请向其索要。

Running Commands

运行命令

All commands use the script at

.claude/skills/mcp-client/scripts/mcp_client.py

bash

undefined

所有命令均使用以下脚本：

.claude/skills/mcp-client/scripts/mcp_client.py

bash

undefined

List configured servers

列出已配置的服务器

python .claude/skills/mcp-client/scripts/mcp_client.py servers

List tools from a server

列出某服务器的工具

python .claude/skills/mcp-client/scripts/mcp_client.py tools <server_name>

Call a tool

调用工具

python .claude/skills/mcp-client/scripts/mcp_client.py call <server> <tool> '{"arg": "value"}'

undefined

python .claude/skills/mcp-client/scripts/mcp_client.py call <server> <tool> '{"arg": "value"}'

undefined

Workflow

工作流程

Check config exists - Run
```
servers
```
command. If error, create
```
.mcp.json
```
List servers - See what MCP servers are configured
List tools - Get tool schemas from a specific server
Call tool - Execute a tool with arguments

检查配置是否存在 - 运行
```
servers
```
命令。若报错，创建
```
.mcp.json
```
列出服务器 - 查看已配置的MCP服务器
列出工具 - 获取特定服务器的工具schema
调用工具 - 使用参数执行工具

Commands Reference

命令参考

Command	Description
`servers`	List all configured MCP servers
`tools <server>`	List tools with full parameter schemas
`call <server> <tool> '<json>'`	Execute a tool with arguments

命令	描述
`servers`	列出所有已配置的MCP服务器
`tools <server>`	列出包含完整参数schema的工具
`call <server> <tool> '<json>'`	使用参数执行工具

Example: Zapier

示例：Zapier

bash

undefined

bash

undefined

1. List servers to confirm Zapier is configured

1. 列出服务器确认Zapier已配置

python .claude/skills/mcp-client/scripts/mcp_client.py servers

2. List Zapier tools

2. 列出Zapier工具

python .claude/skills/mcp-client/scripts/mcp_client.py tools zapier

3. Call a Zapier tool

3. 调用Zapier工具

python .claude/skills/mcp-client/scripts/mcp_client.py call zapier <tool_name> '{"param": "value"}'

undefined

python .claude/skills/mcp-client/scripts/mcp_client.py call zapier <tool_name> '{"param": "value"}'

undefined

Example: Sequential Thinking

示例：Sequential Thinking

bash

undefined

bash

undefined

1. List tools

1. 列出工具

python .claude/skills/mcp-client/scripts/mcp_client.py tools sequential-thinking

2. Use sequential thinking

2. 使用Sequential Thinking

python .claude/skills/mcp-client/scripts/mcp_client.py call sequential-thinking sequentialthinking '{"thought": "Breaking down the problem...", "thoughtNumber": 1, "totalThoughts": 5, "nextThoughtNeeded": true}'

undefined

undefined

Config Format

配置格式

Config file format (

references/mcp-config.json

json

{
  "mcpServers": {
    "zapier": {
      "url": "https://mcp.zapier.com/api/v1/connect",
      "api_key": "your-api-key"
    },
    "sequential-thinking": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-sequential-thinking"]
    }
  }
}

Transport detection:

```
url
```
+
```
api_key
```
→ FastMCP with Bearer auth (Zapier)
```
command
```
+
```
args
```
→ stdio (local servers like sequential-thinking)
```
url
```
ending in
```
/sse
```
→ SSE transport
```
url
```
ending in
```
/mcp
```
→ Streamable HTTP

配置文件格式（

references/mcp-config.json

）：

json

{
  "mcpServers": {
    "zapier": {
      "url": "https://mcp.zapier.com/api/v1/connect",
      "api_key": "your-api-key"
    },
    "sequential-thinking": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-sequential-thinking"]
    }
  }
}

传输方式检测：

```
url
```
+
```
api_key
```
→ 带Bearer认证的FastMCP（Zapier）
```
command
```
+
```
args
```
→ stdio（如sequential-thinking等本地服务器）
以
```
/sse
```
结尾的
```
url
```
→ SSE传输
以
```
/mcp
```
结尾的
```
url
```
→ 可流式HTTP

Error Handling

错误处理

Errors return JSON:

json

{"error": "message", "type": "configuration|validation|connection"}

```
configuration
```
- Config file not found. Create
```
.mcp.json
```
```
validation
```
- Invalid server or tool name
```
connection
```
- Failed to connect to server

错误将返回JSON格式：

json

{"error": "message", "type": "configuration|validation|connection"}

```
configuration
```
- 未找到配置文件。请创建
```
.mcp.json
```
```
validation
```
- 无效的服务器或工具名称
```
connection
```
- 连接服务器失败

Dependencies

依赖

bash

pip install mcp fastmcp

bash

pip install mcp fastmcp

References

参考资料

```
references/example-mcp-config.json
```
- Template config file
```
references/mcp-servers.md
```
- Common server configurations
```
references/python-mcp-sdk.md
```
- Python SDK documentation

```
references/example-mcp-config.json
```
- 配置文件模板
```
references/mcp-servers.md
```
- 常见服务器配置
```
references/python-mcp-sdk.md
```
- Python SDK 文档