搭建全局python 环境

# 安装 venv 工具
sudo  apt install python3.12-venv

# 创建环境
python3 -m venv global_leo_python3_env
# 激活环境
source /home/leo/global_leo_python3_env/bin/activate  # Linux/Mac
myenv\Scripts\activate.bat  # Windows

/home/leo/global_leo_python3_env
❯ tree -L 2
.
├── bin
│   ├── activate
│   ├── activate.csh
│   ├── activate.fish
│   ├── Activate.ps1
│   ├── pip
│   ├── pip3
│   ├── pip3.12
│   ├── python -> python3
│   ├── python3 -> /usr/bin/python3
│   └── python3.12 -> python3
├── include
│   └── python3.12
├── lib
│   └── python3.12
├── lib64 -> lib
└── pyvenv.cfg

7 directories, 11 files

搭建 MCP服务

pip安装：若MCP Server提供Python包，可通过pip install mcp-server安装。

pip install mcp-server

安装完成后，需配置MCP Server，包括环境变量和启动命令。示例如下：

# 设置环境变量
export MCP_SERVER_PORT=8080
export MCP_SERVER_DB_URI=mongodb://localhost:27017/mcp_db

# 启动MCP Server
mcp-server start

vscode 安装Continue

添加模型

增加本地 ollama 模型

/home/leo/.continue/config.yaml

name: Local Config
version: 1.0.0
schema: v1
models:
  - name: Autodetect
    provider: ollama
    model: AUTODETECT

这里可以选择 我们通过 ollama 安装的 模型

配置 mcp 服务

// /home/leo/.continue/config.yaml

mcpServers:
  - uses: anthropic/memory-mcp

安装uv

curl -LsSf https://astral.sh/uv/install.sh | sh


downloading uv 0.9.26 x86_64-unknown-linux-gnu
no checksums to verify
installing to /home/leo/.local/bin
  uv
  uvx
everything's installed!
To add $HOME/.local/bin to your PATH, either restart your shell or run:

# 配置到 .bashrc 中
    source $HOME/.local/bin/env (sh, bash, zsh)
    source $HOME/.local/bin/env.fish (fish)

转载自CSDN-专业IT技术社区

原文链接：https://blog.csdn.net/qq_33689469/article/details/157021804