Lynxe is a Java implementation of Manus, currently used in many applications within Alibaba Group. It is primarily used for handling exploratory tasks that require a certain degree of determinism, such as quickly finding data from massive datasets and converting it into a single row in a database, or analyzing logs and issuing alerts.

You can find some recommended Func-Agent implementations we've prepared at Use Cases.

Lynxe also provides HTTP service invocation capabilities, making it suitable for integration into existing projects. For details, please refer to the developer quick start guide.

A pure Java multi-agent collaboration implementation that provides a complete set of HTTP call interfaces, suitable for secondary integration by Java developers.

Allows you to precisely control every execution detail, providing extremely high execution determinism and completing complex repetitive processes and functions. For specific examples, see Lynxe Use Cases - FuncAgent Use Case

Natively supports the Model Context Protocol (MCP) for seamless integration with external services and tools.

Get Lynxe up and running in under 5 minutes:

• 🌐 DashScope API Key (or alternative AI model provider)

  💡 Get your DashScope API Key: Visit Alibaba Cloud Console, create an API Key in the key management page and copy the key. New users can enjoy 1 million input tokens and 1 million output tokens free quota (valid for 90 days).

• ☕ Java 17+ (for running JAR files or source code execution) or 🐳 Docker (for containerized deployment)

# Download the latest JAR file
wget https://github.com/spring-ai-alibaba/Lynxe/releases/latest/download/lynxe.jar

# Or using curl
curl -L -o lynxe.jar https://github.com/spring-ai-alibaba/Lynxe/releases/latest/download/lynxe.jar

# Run the JAR file
java -jar lynxe.jar

💡 Manual Download: You can also visit the Lynxe Releases page to manually download the latest JAR file.

After the application starts, navigate to http://localhost:18080 in your browser.

💡 Guided Setup: After the application starts, it will automatically display a guided setup page. On the first page, select your language (English/Chinese), then on the second page, enter your DashScope API key to complete the configuration. New users can enjoy 1 million input tokens and 1 million output tokens free quota (valid for 90 days). Visit Alibaba Cloud Console to get your free API key.

🎉 Congratulations! Your multi-agent system has been quickly started. You can visit https://github.com/talk-flow/public-usecase to explore some effective practices we recommend.

# Pull the latest Lynxe Docker image
docker pull ghcr.io/spring-ai-alibaba/lynxe:v4.7.0

# Run the container
docker run -d \
  --name lynxe \
  -p 18080:18080 \
  ghcr.io/spring-ai-alibaba/lynxe:v4.7.0

Run with data persistence (recommended for production):

# Create a directory for data persistence
mkdir -p ./lynxe-data

# Run with volume mounting
docker run -d \
  --name lynxe \
  -p 18080:18080 \
  -v $(pwd)/lynxe-data:/app/data \
  ghcr.io/spring-ai-alibaba/lynxe:v4.7.0

Run with custom environment variables:

docker run -d \
  --name lynxe \
  -p 18080:18080 \
  -e SPRING_PROFILES_ACTIVE=mysql \
  -e SPRING_DATASOURCE_URL=jdbc:mysql://host.docker.internal:3306/lynxe \
  -e SPRING_DATASOURCE_USERNAME=your_username \
  -e SPRING_DATASOURCE_PASSWORD=your_password \
  ghcr.io/spring-ai-alibaba/lynxe:v4.7.0

After the container starts, navigate to http://localhost:18080 in your browser.

💡 Guided Setup: After the application starts, it will automatically display a guided setup page. On the first page, select your language (English/Chinese), then on the second page, enter your DashScope API key to complete the configuration. New users can enjoy 1 million input tokens and 1 million output tokens free quota (valid for 90 days). Visit Alibaba Cloud Console to get your free API key.

# View container logs
docker logs -f lynxe

# Stop the container
docker stop lynxe

# Start the container
docker start lynxe

# Remove the container
docker rm lynxe

🎉 Congratulations! Your multi-agent system is now running in Docker. You can visit https://github.com/talk-flow/public-usecase to explore some effective practices we recommend.

git clone https://github.com/spring-ai-alibaba/Lynxe.git
cd Lynxe

💡 Get your DashScope API Key: Visit Alibaba Cloud Console, create an API Key in the key management page and copy the key. After running the JAR file, access http://localhost:18080 in your browser, and enter your DashScope API key on the guided setup page to complete the configuration. New users can enjoy 1 million input tokens and 1 million output tokens free quota (valid for 90 days).

Using other providers? Update the configuration in src/main/resources/application.yml to use your preferred AI model platform.

Lynxe supports both H2 (default)、MySQL and PostgreSQL databases.

How To Use MySQL/PostgreSQL

1. Configure Database Connection: Update the database configuration and JPA database-platform in the application-mysql.yml/application-postgres.yml under 'src/main/resources/':

   spring:
     datasource:
       url: your_url
       username: your_username
       password: your_password
     jpa:
       database-platform: org.hibernate.dialect.MySQLDialect/PostgreSQLDialect

2. Activate MySQL/PostgreSQL Profile: Update configuration in src/main/resources/application.yml:

   spring:
     ...
     profiles:
       active: mysql/postgres

💡 Note: The application will automatically create required tables on first startup using JPA's ddl-auto: update configuration.

For Unix-like systems (macOS, Linux):

../mvnw spring-boot:run

For Windows systems:

../mvnw.cmd spring-boot:run

Navigate to http://localhost:18080 in your browser.

🎉 Congratulations! Your multi-agent system is now live and ready for action. You can visit https://github.com/Lynxe-public/Lynxe-public-prompts to explore some effective practices we recommend.

you can find stable release from here: release

We enthusiastically welcome contributions from the developer community! Here's how you can make an impact:

You can find available tasks on our project board.

• 🐛 Bug Reports: Submit detailed issue reports
• 💡 Feature Requests: Propose innovative enhancements
• 📝 Documentation: Help us improve clarity and completeness
• 🔧 Code Contributions: Submit pull requests with your improvements

# Fork and clone the repository
git clone [email protected]:spring-ai-alibaba/Lynxe.git
cd Lynxe

# Install project dependencies
mvn clean install

# Apply code formatting standards
mvn spotless:apply

# Start the development server
mvn spring-boot:run

• Follow existing code style and conventions
• Write comprehensive tests for new features
• Update documentation for any API changes
• Ensure all tests pass before submitting PRs

点击这个链接加入钉钉群讨论：钉群链接