Model-Context-Protocol Servers

Name: Model-Context-Protocol Servers
Author: adi2355

@adi2355

代码仓库

关于 Model-Context-Protocol Servers

Collection of purpose-built MCP servers for AI agent workflows.

基本信息

分类

其他

许可证

MIT

运行时

html

传输方式

stdio

发布者

adi2355

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "Model-Context-Protocol-servers": {
      "command": "node",
      "args": [
        "dist/leafly-scraper-cli.js",
        "output.csv",
        "Blue Dream,OG Kush,Sour Diesel"
      ]
    }
  }
}

工具

Scrape a single page with format selection (markdown, HTML, screenshots), custom actions, and content filtering

Discover all URLs on a website and generate a site map

Recursively crawl a website with depth and page limits

Scrape multiple URLs concurrently with queue-based processing

Poll the status of an in-progress batch scrape job

Poll the status of an in-progress crawl job

Search the web and return scraped content from results

LLM-powered structured data extraction using a caller-defined JSON schema

Multi-step research workflow that scrapes, synthesizes, and reports on a topic

Extract standardized cannabis strain data (cannabinoids, terpenes, effects, flavors, interactions)

概览

What is Model-Context-Protocol Servers?

A collection of purpose-built Model Context Protocol (MCP) servers for web scraping and structured data extraction, codebase navigation and analysis, LLM-powered text generation, and JSON querying. Each server exposes its capabilities as MCP tools and resources, making them composable building blocks for AI agent workflows. The servers span TypeScript and Python/JavaScript ecosystems and follow MCP SDK conventions.

How to use Model-Context-Protocol Servers?

Clone the repository, navigate to the desired server directory, install dependencies (npm install for TypeScript/JavaScript servers, pip install for Python), set required environment variables (e.g., FIRECRAWL_API_KEY, DEEPSEEK_API_KEY), and start the server with npm start, node, or python as appropriate. Each server runs on stdio transport by default; the JSON Manager server also supports HTTP transport via the --port flag.

Key features of Model-Context-Protocol Servers

Firecrawl integration for web scraping, crawling, and LLM extraction
Specialized Leafly strain data extractor with regex and LLM strategies
Codebase analysis with file navigation, search, and change monitoring
DeepSeek R1 server for text generation, summarization, and streaming
JSONPath-based querying, filtering, diffing, and result caching
Multi-language support (TypeScript, Python, JavaScript)

Use cases of Model-Context-Protocol Servers

Build AI agents that scrape and extract structured data from websites
Analyze and navigate large codebases programmatically
Generate, summarize, and stream text using DeepSeek AI models
Query, filter, and compare JSON datasets with persistent caching

FAQ from Model-Context-Protocol Servers

What servers are included in this collection?

The collection includes four servers: Firecrawl Web Scraping (TypeScript, stdio), Codebase Analysis (Python, stdio), DeepSeek R1 (JavaScript, stdio), and JSON Manager (JavaScript, stdio/HTTP). Each is documented in the Server Index table.

What runtime dependencies are required?

Node.js for TypeScript and JavaScript servers, Python 3.12+ for the Codebase Analysis server. The Codebase Analysis server also requires the watchdog package for real-time file change monitoring.

How do I set up the Firecrawl Web Scraping server?

Navigate to the firecrawl-mcp-server directory, run npm install, copy .env.example to .env, add your FIRECRAWL_API_KEY, then run npm run build and npm start. Use the CLI command npm run scrape-leafly -- output.csv "strain list" to extract Leafly strain data directly.