Skip to main content
Glama
enesjakozh
ryuichi1208

FastMCP

by ryuichi1208
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

FastMCP——模型上下文协议服务器

使用 FastMCP 实现的轻量级模型上下文协议 (MCP) 服务器， FastMCP是一个用于构建 MCP 服务器和客户端的快速 Pythonic 框架。

特征

  • 创建、检索、更新和删除模型上下文

  • 针对特定上下文执行查询

  • 按模型名称和标签过滤

  • 内存存储（用于开发）

  • FastMCP 集成，轻松实现 MCP 服务器开发

  • 用于指标和监控的 Datadog 集成

Related MCP server: Datadog MCP Server

要求

  • Python 3.7+

  • FastMCP

  • uv（推荐用于环境管理）

  • Datadog 帐户（可选，用于指标）

安装

使用 uv（推荐）

最简单的安装方法是使用提供的脚本：

Unix/Linux/macOS

# Clone the repository
git clone https://github.com/yourusername/datadog-mcp-server.git
cd datadog-mcp-server

# Make the install script executable
chmod +x install.sh

# Run the installer
./install.sh

视窗

# Clone the repository
git clone https://github.com/yourusername/datadog-mcp-server.git
cd datadog-mcp-server

# Run the installer
.\install.ps1

手动安装

# Clone the repository
git clone https://github.com/yourusername/datadog-mcp-server.git
cd datadog-mcp-server

# Create and activate a virtual environment with uv
uv venv
# On Unix/Linux/macOS:
source .venv/bin/activate
# On Windows:
.\.venv\Scripts\activate

# Install dependencies
uv pip install -r requirements.txt

Datadog 配置

该服务器与 Datadog 集成，用于指标和监控。您可以通过多种方式配置 Datadog API 凭据：

1.环境变量

启动服务器之前设置这些环境变量：

# Unix/Linux/macOS
export DATADOG_API_KEY=your_api_key
export DATADOG_APP_KEY=your_app_key  # Optional
export DATADOG_SITE=datadoghq.com    # Optional, default: datadoghq.com

# Windows PowerShell
$env:DATADOG_API_KEY = 'your_api_key'
$env:DATADOG_APP_KEY = 'your_app_key'  # Optional
$env:DATADOG_SITE = 'datadoghq.com'    # Optional

2. .env 文件

在项目目录中创建.env文件：

DATADOG_API_KEY=your_api_key
DATADOG_APP_KEY=your_app_key
DATADOG_SITE=datadoghq.com

3. FastMCP CLI 安装

当作为 Claude Desktop 工具安装时，您可以传递环境变量：

fastmcp install mcp_server.py --name "Model Context Server" -v DATADOG_API_KEY=your_api_key

4.运行时配置

在运行时使用configure_datadog工具：

result = await client.call_tool("configure_datadog", {
    "api_key": "your_api_key",
    "app_key": "your_app_key",  # Optional
    "site": "datadoghq.com"     # Optional
})

用法

启动服务器

# Start directly from activated environment
python mcp_server.py

# Or use uv run (no activation needed)
uv run python mcp_server.py

# Use FastMCP CLI for development (if in activated environment)
fastmcp dev mcp_server.py

# Use FastMCP CLI with uv (no activation needed)
uv run -m fastmcp dev mcp_server.py

安装为 Claude 桌面工具

# From activated environment
fastmcp install mcp_server.py --name "Model Context Server"

# Using uv directly
uv run python -m fastmcp install mcp_server.py --name "Model Context Server"

# With Datadog API key
fastmcp install mcp_server.py --name "Model Context Server" -v DATADOG_API_KEY=your_api_key

使用工具

该服务器提供以下工具：

  • create_context创建一个新的上下文

  • get_context - 检索特定上下文

  • update_context - 更新现有上下文

  • delete_context - 删除上下文

  • list_contexts - 列出所有上下文（带有可选过滤）

  • query_model - 针对特定上下文执行查询

  • health_check - 服务器健康检查

  • configure_datadog - 在运行时配置 Datadog 集成

示例请求

创建上下文

result = await client.call_tool("create_context", {
    "context_id": "model-123",
    "model_name": "gpt-3.5",
    "data": {
        "parameters": {
            "temperature": 0.7
        }
    },
    "tags": ["production", "nlp"]
})

执行查询

result = await client.call_tool("query_model", {
    "context_id": "model-123",
    "query_data": {
        "prompt": "Hello, world!"
    }
})

配置 Datadog

result = await client.call_tool("configure_datadog", {
    "api_key": "your_datadog_api_key",
    "app_key": "your_datadog_app_key",  # Optional
    "site": "datadoghq.com"             # Optional
})

Datadog 指标

服务器向 Datadog 报告以下指标：

  • mcp.contexts.created - 上下文创建事件

  • mcp.contexts.updated - 上下文更新事件

  • mcp.contexts.deleted - 上下文删除事件

  • mcp.contexts.accessed - 上下文访问事件

  • mcp.contexts.total - 上下文总数

  • mcp.contexts.listed - 列出上下文操作事件

  • mcp.queries.executed - 查询执行事件

  • mcp.server.startup - 服务器启动事件

  • mcp.server.shutdown - 服务器关闭事件

发展

请参阅包含的mcp_example.py以获取客户端实现示例：

# Run the example client (with activated environment)
python mcp_example.py

# Run with uv (no activation needed)
uv run python mcp_example.py

执照

麻省理工学院

资源

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • Datadog MCP Server

    brukhabtu
    A
    license
    B
    quality
    Not graded
    maintenance
    A Model Context Protocol server that enables AI assistants to interact with Datadog's observability platform through natural language.
    72
    MIT

  • MCP Python SDK

    hoyoun86
    A
    license
    Not graded
    quality
    D
    maintenance
    A Python implementation of the Model Context Protocol that enables applications to provide standardized context for LLMs, allowing developers to build servers that expose data and functionality to LLM applications.
    MIT

  • MCP Server for Coroot

    jamesbrink
    A
    license
    B
    quality
    B
    maintenance
    A Model Context Protocol server that provides seamless integration with Coroot observability platform, enabling monitoring of applications, analysis of performance metrics, examination of logs and traces, and management of infrastructure through Coroot's comprehensive API.
    61
    1,310 PyPI
    16
    MIT

Appeared in Searches