System prompt with Google Gemini models

This example demonstrates how to set up a Google Gemini model and send a system prompt with Python.

Requirements

• Python 3.12+
• A Google API key

Steps for running this example in the shell

1. Install dependencies:

python -m venv .venv

# macOS/Linux:
source .venv/bin/activate

# Windows Command Prompt:
.venv\Scripts\activate.bat

# Windows PowerShell:
.venv\Scripts\Activate.ps1

pip install -r requirements.txt

2. Export your Google API key as an environment variable:

export GOOGLE_API_KEY="AIza..." # Windows cmd: set GOOGLE_API_KEY="AIza..." # Windows PowerShell: $env:GOOGLE_API_KEY="AIza..."

3. Run the script:

python google-gemini-system-prompt.py

Output

When you run the script, it will send a system prompt and a user prompt to the model, which should provide a response:

=== With system prompt ===
User query: Explain me what is context engineering in simple words
Response: Context engineering involves designing prompts to guide AI models towards desired outputs.

=== With only user prompt ===
User query: Explain me what is context engineering in simple words
Response: Imagine you're talking to a really smart robot. You want it to write a story, but you don't just say "write a story."

**Context engineering is like giving the robot extra information and instructions to help it write the *right* kind of story.**

Think of it like this:

*   **Without context engineering:** You say "Write a story." The robot might write a boring story about a rock.
*   **With context engineering:** You say "Write a *funny* story about a *talking dog* who *tries to become a chef*." Now the robot has a much better idea of what you want!

So, **context engineering is all about carefully crafting the prompts and instructions you give to an AI to get the specific and desired output you're looking for.** It's about providing the right context so the AI understands your needs and can generate something useful and relevant.