About jcrawl4ai-mcp-server
jcrawl4ai-mcp-server is a Java, Spring Boot MCP server for using the Crawl4ai API to crawl URLs and retrieve task results. It is intended for MCP clients that need to start crawls with configurable parameters and fetch their output.
Key features
- 01Crawl specified URLs
- 02Set strategy, depth, and output format
- 03Retrieve results by task ID
Use cases
- 01Crawl websites through Crawl4ai
- 02Fetch results of an earlier crawl
MCP support
It exposes crawl and task operations to MCP clients. The server connects to a configured Crawl4ai API endpoint using its base URL and API token.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEjcrawl4ai-mcp-server
- Java implementation of MCP Server for interacting with Crawl4ai API.
- Certified by mcpreview
Project Overview
jcrawl4ai-mcp-server is a Spring Boot-based MCP server that interacts with the Crawl4ai API to perform web crawling. The main functionalities include:
- Crawling specified URLs using a given strategy, maximum depth, and output format.
- Getting the crawl result by a given task ID.
Configuration
application.properties
Configure the following properties in the src/main/resources/application.properties file:
cawl4ai.base-url: Base URL of the Crawl4ai server.cawl4ai.api-token: API token for the Crawl4ai server.
Example configuration:
cawl4ai.base-url=http://your-cral4ai-server-url:11235
cawl4ai.api-token=your-api-token
Dependencies
The project depends on the following libraries:
- Spring AI MCP Server
- Spring Boot
- Hutool
Running the Project
Build and run the project using Maven:
mvn clean install
java -jar target/jcawl4ai-mcp-server-1.0.0.jar
You can download the jar file from this link directly.
APIs
Crawl4aiApi
crawl Method
- Description: Call the Crawl4ai API to crawl the specified URLs.
- Parameters:
urls: Array of target website URLs.strategy: Crawl strategy.max_depth: Maximum depth.output_format: Output format.
- Return Value: JSON string of the crawl result.
task Method
- Description: Get the crawl result by a given task ID.
- Parameters:
taskId: Task ID.
- Return Value: JSON string of the crawl result.
Logging
Log file path: ./target/mcp-stdio-server.log.
MCP Server Configuration
{
"mcpServers": {
"jcawl4ai-mcp-server": {
"autoApprove": [
"crawl",
"task"
],
"disabled": false,
"timeout": 60,
"command": "java",
"args": [
"-jar",
"/path/to/your/jar/file/jcawl4ai-mcp-server-1.0.0.jar"
],
"transportType": "stdio"
}
}
}
Contact
If you have any questions or suggestions, please contact Ken Ye.
jcrawl4ai-mcp-server
Java 实现的 MCP 服务器,用于与 Crawl4ai API 进行交互。
项目概述
jcrawl4ai-mcp-server 是一个基于 Spring Boot 的 MCP 服务器,用于调用 Crawl4ai API 进行网页爬取。该项目的主要功能包括:
- 使用指定的策略、最大深度和输出格式对给定的 URL 进行爬取。
- 根据给定的任务 ID 获取爬取结果。
配置
application.properties
在 src/main/resources/application.properties 文件中配置以下属性:
cawl4ai.base-url:Crawl4ai 服务器的基础 URL。cawl4ai.api-token:Crawl4ai 服务器的 API 令牌。
示例配置:
cawl4ai.base-url=http://your-cral4ai-server-url:11235
cawl4ai.api-token=your-api-token
依赖
项目依赖于以下库:
- Spring AI MCP Server
- Spring Boot
- Hutool
启动
使用 Maven 构建并运行项目:
mvn clean install
java -jar target/jcawl4ai-mcp-server-1.0.0.jar
您可以从以下链接中直接下载jar包: link
接口
Crawl4aiApi
crawl 方法
- 描述:调用 Crawl4ai API 爬取指定的 URL。
- 参数:
urls:目标网站的 URL 数组。strategy:爬取策略。max_depth:最大深度。output_format:输出格式。
- 返回值:爬取结果的 JSON 字符串。
task 方法
- 描述:根据给定的任务 ID 获取爬取结果。
- 参数:
taskId:任务 ID。
- 返回值:爬取结果的 JSON 字符串。
日志
日志文件路径为 ./target/mcp-stdio-server.log。
MCP Server 配置
{
"mcpServers": {
"jcawl4ai-mcp-server": {
"autoApprove": [
"crawl",
"task"
],
"disabled": false,
"timeout": 60,
"command": "java",
"args": [
"-jar",
"/path/to/your/jar/file/jcawl4ai-mcp-server-1.0.0.jar"
],
"transportType": "stdio"
}
}
}
联系
如果您有任何问题或建议,请联系 Ken Ye。
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.