mcp-server-requests

Name: mcp-server-requests
Author: MCP-Mirror

@MCP-Mirror

代码仓库

关于 mcp-server-requests

Mirror of

#mcp-server-requests #http-requests #llm-integration

基本信息

分类

其他

许可证

MIT license

运行时

python

传输方式

stdio

发布者

MCP-Mirror

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "coucya_mcp-server-requests": {
      "command": "python",
      "args": [
        "-m",
        "mcp_server_requests"
      ]
    }
  }
}

工具

**url** (required): Target URL

Same as http_post

概览

What is mcp-server-requests?

mcp-server-requests is an MCP server that provides HTTP request capabilities, enabling LLMs to fetch and process web content. It is designed for developers and AI applications that need programmatic web fetching and API interaction.

How to use mcp-server-requests?

Install by cloning the repository and running pip install .. Configure in the MCP client by adding "mcp-server-requests" with command python -m mcp_server_requests. Optionally set User-Agent via command-line flags (--user-agent or --random-user-agent). Use the server’s tools (fetch, http_get, http_post, http_put, http_patch, http_delete) to perform HTTP operations.

Key features of mcp-server-requests

Converts web content to Markdown format
Filters out content useless for LLMs (e.g., scripts, styles)
Supports custom User-Agent headers
Supports randomly generated User-Agent headers
Supports custom request headers in HTTP calls
Full HTTP methods: GET, POST, PUT, DELETE, PATCH

Use cases of mcp-server-requests

Fetching web pages and converting them to clean Markdown for LLMs
Sending API requests (GET, POST, PUT, etc.) with custom headers
Retrieving and processing HTTP response headers programmatically
Testing or prototyping HTTP interactions within an MCP environment

FAQ from mcp-server-requests

What runtime does mcp-server-requests require?

Python is required. Install via pip install . after cloning the repository.

How do I set a custom User-Agent?

Use the --user-agent flag with a string, or --random-user-agent to generate one randomly. The --force-user-agent flag controls priority over LLM-provided User-Agents.

What content filtering options are available?

The fetch tool accepts return_content with values: raw, basic_clean, strict_clean, or markdown. These control how much HTML is stripped before returning.