# Ask AI

# Introduction

Armada API documentation is designed to be AI-friendly, providing structured content that can be easily consumed by Large Language Models (LLMs) to assist you in your integration journey. We offer a dedicated llms.txt file that contains the complete documentation in a format optimized for AI processing.


# How to Use

Point your AI assistant directly to our documentation URL:

URL
https://docs.armadadelivery.com/v1/static/llms.txt

# Recommended Usage Prompt

For the best results when using AI to understand our API, we recommend using this prompt template:

Prompt
Use https://docs.armadadelivery.com/v1/static/llms.txt to answer anything related to API integration with Armada.
Do not use any other external resources or make assumptions on your own.
If the answer is not found in the llms.txt or is out of context, do not provide a random response.

Example use cases:

  • "Help me implement the HMAC signature generation for app installation"
  • "Explain how the authentication works"
  • "Explain how to handle the order.updated webhook"
  • "Show me how to create an order for Kuwait with the API"
  • "What are the required fields for branch configuration?"
  • "What is the best way for my customers to track their orders?"
  • "Show me how to create order without branch id"
  • "How my customers will handle their minimum wallet balance?"

# Recommended AI Tool

For optimal results, we recommend using Claude by Anthropic.
We found that Claude stays very well within the provided context and does not give incorrect or hallucinated answers.

Other AI assistants like ChatGPT, Gemini, or GitHub Copilot can also be effective when provided with our documentation, But Claude is highly recommend.


# Important Disclaimer


# Best Practices

Always cross-reference AI-generated responses with the official documentation. If something seems unclear or contradictory, check the actual docs.

When asking AI for help, provide specific context about what you're trying to achieve, your programming language, and any error messages you're encountering.

AI-generated code should be tested in a development environment with your developer accounts before production use.

Instead of "How does the API work?", ask targeted questions like "How do I validate the callback webhook signature?"

The llms.txt file is regularly updated. If you've downloaded it, check back periodically for the latest version.