BIP on AI Agents

During this BIP you will use TOMMI Lite, a lightweight framework for building and testing AI agents. This page contains everything BIP participants need to prepare before the 15th of June. Please complete the following tasks:

1. Learn to use a terminal — Read through Section 1 below and practise opening a terminal on your computer, navigating folders, and running basic commands.
2. Create a free* ( New) or payable Mistral account — Go to console.mistral.ai and sign up. You can now choose between a free Experiment plan or a paid plan. Both work with TOMMI Lite. See Section 1b below for detailed instructions on both options. Generate an API key and keep it at hand for the next step.*

* Note: During the virtual session we were not aware that MISTRAL offered also free accounts. Apparently these free accounts should be enough for you to complete this BIP satisfactorily. Sorry about the confusion and see you soon in Málaga! 

3. Download and install TOMMI Lite — Download the zip for your operating system (Section 2), extract it, configure your API key in env.txt, and start the server following the instructions in Section 3.
4. Test at least one or two AI agents — Once the server is running, open http://localhost:8000 in your browser, select an agent from the sidebar (e.g. Hello World), and have a short conversation to confirm everything works.

Before You Arrive: What to Bring

For the in-person component of the BIP, please make sure you have the following ready before you arrive in Málaga:

Equipment

• A laptop with Wi-Fi access — you will be working on your own machine throughout all sessions.
• A Mistral account (a free account is sufficient — see Section 1b) with a working API key configured in TOMMI Lite.

Data for Your Own Agent (optional but recommended)

If you would like to go beyond the provided examples and build one or more agents that you can actually use in your work afterwards, you will need to bring the relevant data with you. The type of data depends on the task you have in mind. Here are some examples:

1. Terminal Basics

A terminal (also called "command line" or "console") is a text-based interface where you type commands to interact with your computer. You will need it to run TOMMI Lite and manage your AI agents.

macOS

# Navigate to the TOMMI Lite folder
cd ~/Downloads/tommilite

# Start the server
./start.sh

Windows

:: Navigate to the TOMMI Lite folder
cd C:\Users\YourName\Downloads\tommilite

:: Start the server
start.bat

Common Tips

• Tab completion: Press Tab while typing a folder or file name to auto-complete it.
• Command history: Press the Up arrow key to recall previous commands.
• Copy/Paste: On Mac, use Cmd+C / Cmd+V. On Windows terminal, use Ctrl+Shift+C / Ctrl+Shift+V (or right-click to paste).
• If something goes wrong: Press Ctrl + C to stop whatever is running, then try again.

New 1b. Create Your Mistral Account

Step 1: Sign Up

1. Go to Mistral AI Studio.
2. Click Sign Up and fill in your details.
3. Verify your email address.

Step 2: Choose Your Plan

Once logged in, navigate to the Subscription section. You have two options:

Step 3: Generate an API Key

1. In Mistral AI Studio, go to the API Keys section.
2. Click Create New Key.
3. Give your key a name (e.g., "AI Agent Project").
4. Copy the generated API key and store it securely.

Step 4: Add the Key to TOMMI Lite

Open the file env.txt in your tommi_lite folder with any text editor and paste your key:

# LLM Provider configuration
LLM_PROVIDER=mistral
MISTRAL_API_KEY=your_key_here
MISTRAL_MODEL=mistral-small-latest

Replace paste_your_key_here with the actual API key you copied in Step 3.

Step 5: Monitor Usage

Check your usage and limits at any time in the Limits section of Mistral AI Studio.

2. Download TOMMI Lite

TOMMI Lite is the lightweight framework you will use to build and test AI agents during the BIP. Download the version for your operating system:

3. Setting Up TOMMI Lite

Prerequisites

3.1 Choose a Large Language Model (LLM)

TOMMI Lite needs an LLM to power its agents. You have two options — choose one:

3.2 Extract TOMMI Lite

Unzip the downloaded file to any folder on your computer. You will get a folder called tommi_lite containing all the necessary files.

# Navigate to where you downloaded the zip
cd ~/Downloads
unzip tommi_lite.zip
cd tommi_lite

:: Right-click tommi_lite.zip and select "Extract All"
:: Then open a Command Prompt and navigate to the folder:
cd %USERPROFILE%\Downloads\tommi_lite

3.3 Verify the folder structure

You should see these files:

tommi_lite/
  app.py              # Server
  agent_runner.py     # Agent engine
  llm_client.py       # LLM provider abstraction
  env.txt             # Configuration
  requirements.txt    # Python dependencies
  start.sh            # Mac/Linux startup
  start.bat           # Windows startup
  static/
    index.html        # Chat interface
    create.html       # Agent creator
  agents/
    hello_world/      # Example agent

3.4 Configure your API Key (Mistral Cloud only)

If you chose Mistral Cloud, open the file env.txt in the tommilite folder with any text editor:

# LLM Provider configuration
LLM_PROVIDER=mistral
MISTRAL_API_KEY=YOUR_MISTRAL_API_HERE
MISTRAL_MODEL=mistral-small-latest

Replace YOUR_MISTRAL_API_HERE with your actual Mistral API key (obtained from console.mistral.ai).

3.5 Start the Server

The start script automatically creates a Python virtual environment, installs dependencies, and starts the server.

./start.sh

:: Double-click start.bat, or from Command Prompt:
start.bat

If Python is not installed, the script will ask you to accept its installation. Type Y (or just press Enter) to proceed.

The first time, it will also install dependencies (this takes about 30 seconds). You should see:

  TOMMI Lite running at http://localhost:8000

3.6 Open the Interface and Chat with an Agent

Open your browser and go to:

http://localhost:8000

You will see the TOMMI Lite interface, which has three main areas:

• Sidebar (left) — Lists all available agents. You will see Hello World Agent pre-installed as an example.
• Chat area (centre) — Where conversations with your agent take place. Messages appear here in real time as the agent responds.
• Input bar (bottom) — Type your message and press Enter or click Send.

To start chatting:

1. Click on an agent in the sidebar (e.g. Hello World) to select it.
2. The agent's welcome message will appear in the chat area, along with example queries you can click.
3. Type a message in the input bar and press Enter. The agent will stream its response in real time.
4. To start a fresh conversation, click the "New Chat" button in the top bar.

4. Troubleshooting

Here are the most common issues encountered during setup and how to fix them:

chmod +x start.sh