TangramCollaborativeAI

Launching an Agent Server

This guide will walk you through setting up a Python environment, installing dependencies, and launching an agent server in the TangramCollaborativeAI project.

---

🔹 Step 1: Create a Python Virtual Environment

To ensure a clean workspace, it’s best to create a Python virtual environment.

1. Open a terminal and navigate to the agent’s directory:

   cd models/[agent_directory]
   

   Replace [agent_directory] with the actual agent’s folder name.

2. Create a virtual environment inside the agent’s directory:

   python -m venv venv
   

3. Activate the virtual environment:
   • On macOS/Linux:

     source venv/bin/activate
     

   • On Windows (CMD):

     venv\Scripts\activate
     

   • On Windows (PowerShell):

     venv\Scripts\Activate.ps1
     

---

🔹 Step 2: Install Requirements

Each agent has its own dependencies listed in a requirements.txt file inside its directory. Once the virtual environment is activated, install them with:

pip install -r requirements.txt

To verify the installation, run:

pip list

---

🔹 Step 3: Environment Variables

The current implemented agents simpleGPTAgent and relationalAgentGPT expect an “OPENAI_API_KEY” Environment Variable to exist in the system, and to contain contain a valid openAI key.

---

🔹 Step 4: Launch the Agent Server

To start the agent server, you must run the command from the TangramCollaborativeAI directory (one level above models).

1. Navigate to the project root:

   cd ../../  # Go to TangramCollaborativeAI
   

2. Run the agent server using the following command structure:

   python -m models.[agent_directory].[agent_main_file]
   

   Replace:

   • [agent_directory] with the agent’s folder name.
   • [agent_main_file] with the main script name (without .py).

Available Agents

There are currently 3 Agents available to use, our most complex agent that utilizes spacial descriptions relationalAgent, our self reflective agent simpleGPTAgent and a random playing agent corresponding to the default template behaviour

Example:

To launch the RelationalAgent, use:

python -m models.relationalAgent.relationalAgentGPT

To launch the SimpleGPTAgent, use:

python -m models.simpleAgent.simpleGPTAgent

To launch the class default random behaviour, use:

python -m models.templates.agentTemplate

---

Now your agent server should be up and running!

This site is open source. Improve this page.