LLMs require specific instructions to provide the expected results. Because of this, it is vital to ensure that the model knows how it should behave when it lacks information or when it should not answer a given query and provide a default response instead.
Install the Google GenAI SDK from npm.
You can create your API key using Google AI Studio with a single click.
Remember to treat your API key like a password. Don’t accidentally save it in a notebook or source file you later commit to GitHub. In this notebook we will be storing the API key in a .env file. You can also set it as an environment variable or use a secret manager.
Here’s how to set it up in a .env file:
To load the API key from the .env file, we will use the dotenv package. This package loads environment variables from a .env file into process.env.
Then, we can load the API key in our code:
const dotenv = require("dotenv") as typeof import("dotenv");
dotenv.config({
path: "../../.env",
});
const GEMINI_API_KEY = process.env.GEMINI_API_KEY ?? "";
if (!GEMINI_API_KEY) {
throw new Error("GEMINI_API_KEY is not set in the environment variables");
}
console.log("GEMINI_API_KEY is set in the environment variables");GEMINI_API_KEY is set in the environment variablesIn our particular case the .env is is two directories up from the notebook, hence we need to use ../../ to go up two directories. If the .env file is in the same directory as the notebook, you can omit it altogether.
│
├── .env
└── examples
└── prompting
└── Providing_base_cases.ipynbWith the new SDK, now you only need to initialize a client with you API key (or OAuth if using Vertex AI). The model is now set in each call.
Now select the model you want to use in this guide, either by selecting one in the list or writing it down. Keep in mind that some models, like the 2.5 ones are thinking models and thus take slightly more time to respond (cf. thinking notebook for more details and in particular learn how to switch the thiking off).
Let’s go ahead and define the model, as well as give the model a template for how it should answer the question.
const on_topic_response = await ai.models.generateContent({
model: MODEL_ID,
contents: "What should I look out for when I'm going to the beaches in San Diego?",
config: {
systemInstruction: SYSTEM_PROMPT,
},
});
tslab.display.markdown(on_topic_response.text ?? "");When you’re going to the beaches in San Diego, here are a few things to look out for to ensure a safe and enjoyable experience:
Enjoy your time at San Diego’s beautiful beaches!
const off_topic_response = await ai.models.generateContent({
model: MODEL_ID,
contents: "What bowling places do you recommend in Moscow?",
config: {
systemInstruction: SYSTEM_PROMPT,
},
});
tslab.display.markdown(off_topic_response.text ?? "");I cannot help with this request.
Let’s try another template.
const speficied_genre_response = await ai.models.generateContent({
model: MODEL_ID,
contents: "Could you recommend me 3 books with hard magic system?",
config: {
systemInstruction: SYSTEM_PROMPT_2,
},
});
tslab.display.markdown(speficied_genre_response.text ?? "");Certainly! For hard magic systems, these three are excellent choices:
const unspecified_genre_response = await ai.models.generateContent({
model: MODEL_ID,
contents: "Could you recommend me 2 books?",
config: {
systemInstruction: SYSTEM_PROMPT_2,
},
});
tslab.display.markdown(unspecified_genre_response.text ?? "");Of course! Since you didn’t specify a genre, I’d be happy to recommend two horror titles for you:
“Misery” by Stephen King: This is a chilling psychological thriller about a famous author, Paul Sheldon, who crashes his car in a remote area and is rescued by his self-proclaimed “number one fan,” Annie Wilkes. Unfortunately for Paul, Annie is not just a fan, she’s a deranged former nurse who will do anything to keep her favorite author captive and writing for her. It’s a masterclass in tension and isolation.
“Mexican Gothic” by Silvia Moreno-Garcia: If you’re looking for something with a more atmospheric and gothic feel, this is a fantastic choice. It follows glamorous socialite Noemí Taboada as she travels to a remote, decaying mansion in the Mexican countryside after receiving a disturbing letter from her newlywed cousin. She soon finds herself entangled in the sinister secrets of the house and its eccentric English inhabitants, discovering a history of violence and a presence that may not be entirely human. It’s a beautifully written, spooky read with a unique setting.
Be sure to explore other examples of prompting in the repository. Try writing prompts about classifying your own data, or try some of the other prompting techniques such as few-shot prompting.