Build multi-agent systems with OS reusable agents

So in this tutorial, we are going to go over a few things:

• Finding agents for your use case
• Running the agents on Coral server
• Having them all collaborate together

​

1. Choose your agents

We go over this step in the agent discovery part of our docs so you might want to look there if you have not but in short you can choose the agents you want to use for your multi-agent system from this repo: https://github.com/Coral-Protocol/awesome-agents-for-multi-agent-systems

​

2. Set them up on the Coral server

Now you have your agents, let’s run and set each of them up individually

​

3. Have your agents collaborate

​

3.1 Clone Coral Protocol and Interface Agent

First, clone required repositories:

git clone https://github.com/Coral-Protocol/coral-server.git
git clone https://github.com/Coral-Protocol/Coral-Interface-Agent.git

​

3.2 Set up Interface Agent

• Install Dependencies:

  Install all required packages:

  Copy

  pip install langchain-mcp-adapters==0.0.10 langchain-openai langchain langchain-core langchain-community

• Configure Environment Variables:

  Copy

  export OPENAI_API_KEY=sk-xxx

​

3.3 Start the MCP Server

Navigate to the coral-server directory and run:

cd coral-server
./gradlew run

Note: Gradle may appear to stall at 83%, but the server is running. Check the terminal logs to confirm.

​

3.4 Run your chosen agents in separate terminals

For example, if you choose the Coral GitClone Agent, Coral CodeDiffReview Agent, and Coral RepoUnitTestAdvisor Agent, please run these three agents along with the Interface Agent in four separate terminals:

# Terminal 1: Interface Agent
python 0-langchain-interface.py

# Terminal 2: GitClone Agent
python 1-crewai-GitCloneAgent.py

# Terminal 3: CodeDiffReview Agent
python 2-camel-CodeDiffReviewAgent.py

# Terminal 4: UnitTestRunner Agent
python 3-langchain-UnitTestRunnerAgent.py

​

Thank you

• For issues, please open a GitHub issue in the corresponding agent repository.