Website Scraper MCP Server
Enables AI agents to scrape, crawl, clean, chunk, and index website content into Azure AI Search for full-text retrieval.
README
Website Scraper MCP Server
A production-ready MCP (Model Context Protocol) server that allows any MCP-compatible AI agent to scrape websites, crawl internal pages, clean content, chunk it, and index everything into Azure AI Search — all through a clean, typed tool interface.
Table of Contents
- Architecture
- Tools
- Installation
- Running Locally
- Running with Docker
- Environment Variables
- Sample MCP Client
- Example API Requests
Architecture
website_scraper_mcp/
├── app.py ← Entry point (stdio / SSE transport)
├── server.py ← MCP server + tool dispatcher
├── config.py ← Pydantic Settings (env vars)
├── models.py ← Input/Output Pydantic models
└── tools/
├── scrape.py ← Tool 1 – static/dynamic detection + scraping
├── crawl.py ← Tool 2 – BFS crawler, robots.txt aware
├── clean.py ← Tool 3 – Trafilatura + BS4 content cleaning
├── chunk.py ← Tool 4 – sliding window chunking
└── azure_ai_search.py ← Tools 5 & 7 – index + search
Tools
| # | Tool | Description |
|---|---|---|
| 1 | scrape_website |
Detect static/dynamic, scrape title/content/links |
| 2 | crawl_website |
BFS crawl with depth limit + robots.txt |
| 3 | clean_content |
Strip noise HTML, return readable text |
| 4 | chunk_content |
Sliding window chunks (~1 000 chars, 200 overlap) |
| 5 | index_to_ai_search |
Upload chunks to Azure AI Search |
| 6 | index_website |
End-to-end pipeline: crawl → clean → chunk → index |
| 7 | search_index |
Full-text search on the Azure AI Search index |
Installation
Prerequisites
- Python 3.11+
- Azure AI Search service (free tier works for testing)
Steps
# 1. Clone the repo
git clone https://github.com/your-org/website-scraper-mcp.git
cd website-scraper-mcp
# 2. Create and activate a virtual environment
python -m venv .venv
# Windows
.venv\Scripts\activate
# Linux/Mac
source .venv/bin/activate
# 3. Install dependencies
pip install -r requirements.txt
# 4. Install Playwright browsers (Chromium)
playwright install chromium
# 5. Copy and fill environment variables
cp .env.example .env
# Edit .env with your Azure credentials
Running Locally
stdio mode (default — for MCP clients / AI agents)
python -m website_scraper_mcp.app
# or
python -m website_scraper_mcp.app --transport stdio
SSE mode (HTTP endpoint for browser-based / HTTP clients)
python -m website_scraper_mcp.app --transport sse --port 8000
# Server available at http://localhost:8000/sse
Running with Docker
# Build and start in SSE mode
docker compose up --build
# Stop
docker compose down
The container exposes port 8000 for SSE transport.
Environment Variables
| Variable | Default | Description |
|---|---|---|
AZURE_SEARCH_ENDPOINT |
(required) | Azure AI Search service URL |
AZURE_SEARCH_KEY |
(required) | Admin API key |
AZURE_SEARCH_INDEX_NAME |
website-content |
Target index name |
PLAYWRIGHT_TIMEOUT_MS |
30000 |
Playwright page load timeout (ms) |
PLAYWRIGHT_HEADLESS |
true |
Run Chromium headless |
MAX_CRAWL_DEPTH |
2 |
Maximum crawl depth |
MAX_PAGES_PER_SITE |
100 |
Hard cap on pages per crawl |
CRAWL_DELAY_SECONDS |
0.5 |
Polite delay between requests |
CHUNK_SIZE |
1000 |
Characters per chunk |
CHUNK_OVERLAP |
200 |
Overlap between consecutive chunks |
LOG_LEVEL |
INFO |
Python logging level |
Sample MCP Client
Run the included example after starting the server in SSE mode:
python examples/mcp_client_example.py
Or configure it in your MCP-compatible agent (e.g. Claude Desktop mcp_config.json):
{
"mcpServers": {
"website-scraper": {
"command": "python",
"args": ["-m", "website_scraper_mcp.app", "--transport", "stdio"],
"cwd": "/path/to/website-scraper-mcp",
"env": {
"AZURE_SEARCH_ENDPOINT": "https://your-service.search.windows.net",
"AZURE_SEARCH_KEY": "your-key",
"AZURE_SEARCH_INDEX_NAME": "website-content"
}
}
}
}
Example API Requests
Via MCP client (Python SDK)
import asyncio
from mcp import ClientSession
from mcp.client.sse import sse_client
async def demo():
async with sse_client("http://localhost:8000/sse") as (read, write):
async with ClientSession(read, write) as session:
await session.initialize()
# Scrape a single page
result = await session.call_tool("scrape_website", {"url": "https://example.com"})
print(result)
# Full pipeline
result = await session.call_tool("index_website", {
"url": "https://example.com",
"max_depth": 2
})
print(result)
# Search
result = await session.call_tool("search_index", {
"query": "What services does the company provide?",
"top": 5
})
print(result)
asyncio.run(demo())
Tool input/output examples
scrape_website
// Input
{"url": "https://example.com"}
// Output
{
"title": "Example Domain",
"url": "https://example.com",
"content": "This domain is for use in illustrative examples...",
"links": ["https://www.iana.org/domains/example"],
"is_dynamic": false,
"metadata": {"description": "..."}
}
index_website
// Input
{"url": "https://example.com", "max_depth": 2}
// Output
{
"url": "https://example.com",
"pages_crawled": 4,
"total_chunks": 38,
"indexed_documents": 38,
"failed_documents": 0,
"status": "success",
"errors": []
}
search_index
// Input
{"query": "What services does the company provide?", "top": 5}
// Output
{
"query": "What services does the company provide?",
"total_results": 3,
"hits": [
{
"id": "abc123",
"url": "https://example.com/services",
"title": "Our Services",
"content": "We provide cloud, AI, and data services...",
"chunk_number": 0,
"score": 9.8
}
]
}
Error Handling
The server handles all errors gracefully and returns structured JSON error responses:
{
"error": "HTTP 404 when fetching https://example.com/missing",
"tool": "scrape_website"
}
Handled errors include: invalid URLs, HTTP 4xx/5xx, timeouts, Playwright failures, Azure Search quota errors, network issues, and duplicate document IDs.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.