
CopilotMCP
A collection of MCP servers built with FastMCP framework that handle various tasks including customer interviews, E2E testing, and go-live processes, enabling seamless integration with GitHub Copilot through VSCode.
README
MCP Server Project
This project contains multiple MCP servers for handling various tasks such as customer interviews, E2E testing, and go-live processes. Each MCP server is implemented using the FastMCP framework.
Prerequisites
- Python 3.11 or higher
uv
package manager (used instead ofpip
)
Setup
-
Clone the repository:
git clone <repository-url> cd CopilotMCP
-
Install dependencies using
uv
:uv install
Running the Project
To run a specific MCP server, use the following command:
uv run main.py --mcp <mcp_server_name>
Replace <mcp_server_name>
with the name of the MCP server you want to run. Available options are:
hello
customer_mcp
interview_mcp
go_live_mcp
testing_e2e_mcp
Example
To run the customer_mcp
server:
uv run main.py --mcp customer_mcp
This will start the customer_mcp
server and make it accessible at:
http://127.0.0.1:8000/customer-mcp-server/mcp
Running with Docker Compose
You can use Docker Compose to run all MCP servers simultaneously. Each server will be exposed on a different port.
Steps to Run
-
Build and start all services:
docker-compose up --build
-
Access the MCP servers at the following URLs:
hello-mcp
: http://127.0.0.1:8000/hello-server/mcpcustomer-mcp
: http://127.0.0.1:8001/customer-mcp-server/mcpinterview-mcp
: http://127.0.0.1:8002/interview-mcp-server/mcpgo-live-mcp
: http://127.0.0.1:8003/go-live-mcp-server/mcptesting-e2e-mcp
: http://127.0.0.1:8004/testing-e2e-mcp-server/mcp
-
Stop all services:
docker-compose down
Notes
- Ensure Docker and Docker Compose are installed on your system.
- Each MCP server runs in its own container and is accessible on its respective port.
Configuring MCP Servers in VSCode
To configure MCP servers in VSCode, you can add entries to your settings.json
file under the mcp.servers
section. This allows you to define and manage MCP server endpoints for easy access.
Example Configuration
To configure the hello-mcp-server
, add the following entry to your settings.json
file:
"hello-mcp-server": {
"url": "http://127.0.0.1:8000/hello-server/mcp/http",
"type": "http"
}
Then you can access hello server hello
tool by executing: #hello <something>
in Copilot.
Steps to Add Configuration
- Open your VSCode
settings.json
file. - Locate or create the
mcp.servers
section. - Add the configuration for the desired MCP server, as shown in the example above.
Accessing the Server
Once configured, you can use the defined URL to interact with the MCP server. For example, the hello-mcp-server
will be accessible at:
http://127.0.0.1:8000/hello-server/mcp/http
This setup ensures that you can easily manage and test MCP servers directly from VSCode.
Project Structure
main.py
: Entry point for running MCP servers.customer_mcp.py
: Handles customer interview-related tasks.interview_mcp.py
: Manages customer interview steps.go_live_mcp.py
: Handles go-live processes.testing_e2e_mcp.py
: Manages E2E testing tasks.hello.py
: Example MCP server for testing.
Notes
- Ensure that the
uv
package manager is installed and configured correctly. - Use the
--mcp
argument to specify which MCP server to run.
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。