[Ask GPT for the CLI] Bash script that calls the Azure OpenAI API to ask a question and returns the answer #openai #gpt

README.md

ask_gpt.sh

A script to call the Azure OpenAI API and ask a question, returning the answer.

Description

ask_gpt.sh is a Bash script that interacts with the Azure OpenAI API to ask a question and retrieve the corresponding answer. It simplifies the process of making API calls and handling the JSON response.

Usage

./ask_gpt.sh "What is the meaning of life?"

Replace "What is the meaning of life?" with your desired question.

Requirements

• The jq command line tool must be installed.
• The following environment variables must be set:
  • AZURE_OPENAI_API_KEY: Azure OpenAI API key.
  • AZURE_OPENAI_ENDPOINT: Azure OpenAI endpoint.
  • OPENAI_API_VERSION: Azure OpenAI API version.
  • CHAT_DEPLOYMENT: Azure OpenAI deployment name.

Installation

1. Clone the repository:

   git clone <repository_url>

2. Change into the repository directory:

   cd ask-gpt

3. Make the script executable:

   chmod +x ask_gpt.sh

Execution

Run the script with your desired question as an argument:

./ask_gpt.sh "What is the meaning of life?"

Output

The script sends a request to the Azure OpenAI API, passing the question as a user message. It then retrieves the answer from the API response and displays it in the console.

License

This project is licensed under the MIT License.

ask_gpt.sh

#!/usr/bin/env bash
#
# ask_gpt.sh
# Usage: ./ask_gpt.sh "What is the meaning of life?"
#
# Description: This script calls the Azure OpenAI API to ask a question and returns the answer.
#
# Arguments:
#   - The question is passed as the first argument to the script.
#
# Requirements:
#   - The jq command line tool must be installed.
#   - The Azure OpenAI API key must be set in the AZURE_OPENAI_API_KEY environment variable.
#   - The Azure OpenAI endpoint must be set in the AZURE_OPENAI_ENDPOINT environment variable.
#   - The Azure OpenAI API version must be set in the OPENAI_API_VERSION environment variable.
#   - The Azure OpenAI deployment name must be set in the CHAT_DEPLOYMENT environment variable.

# Make API call to Azure OpenAI API and print the answer
curl -s "$AZURE_OPENAI_ENDPOINT/openai/deployments/$CHAT_DEPLOYMENT/chat/completions?api-version=$OPENAI_API_VERSION" \
  -H "Content-Type: application/json" \
  -H "api-key: $AZURE_OPENAI_API_KEY" \
  -d "{\"messages\":[{\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},{\"role\": \"user\", \"content\": \"$1\"}]}" \
  | jq '.choices | .[0] | .message | .content' \
  | echo -e "$(cat -)"