Apimatic CLI 🚀

A Node.js CLI wrapper for the Apimatic Python package.
It allows you to run Apimatic commands directly from Node / npm while leveraging the underlying Python engine.

Here’s a cleaned-up and more beginner-friendly version of your installation guide:

📦 Installation Guide

1️⃣ Install the CLI (NPM)

# Install globally via npm
npm install -g apimatic-cli

2️⃣ Install the Python Package (PyPI)

# Install globally via pip
pip install Apimatic

⚠️ Note:

• This CLI is just an NPM wrapper around the Python package.
• You must install both for it to work correctly.

✅ Verify Installation

# Check if apimatic-cli is installed globally
npm list -g --depth=0

# Check if Apimatic is installed globally in Python (avoid virtualenvs)
pip list

You should see both apimatic-cli (npm) and Apimatic (pip) listed.

🔄 Upgrade to Latest Version

# Update the npm wrapper
npm update -g apimatic-cli

# Upgrade the Python package
pip install --upgrade Apimatic

Would you like me to also make a diagram/flowchart image showing how apimatic-cli (npm) just calls Apimatic (Python) so users instantly understand the relationship?

⚡ Usage

Apimatic now uses commands to separate actions.

1. Generate Documentation:

apimatic generate [OPTIONS]

2. Configure API Keys:

apimatic config [OPTIONS]

This internally calls the Python module: python3 -m Apimatic.cli [command] [options]

🔑 Configuration

To use AI enhancements from providers like OpenAI, Google Gemini, or Groq, you must set an API key.

# Set your OpenAI API key
apimatic config --set-openai-key YOUR_API_KEY

# Set your Google Gemini API key
apimatic config --set-gemini-key YOUR_API_KEY

# Set your Groq API key
apimatic config --set-groq-key YOUR_API_KEY

The key will be stored securely in your home directory for the Python package to use.

⚙️ Generation Options

| Option | Description | | --- | --- | | -h, --help | Show help message and exit | | --src SRC | Root directory of the project to scan (Default: current directory) | | --framework [FRAMEWORK ...] | Force a specific framework (flask, fastapi, etc.). If omitted, auto-detected | | --format {markdown} | Output format (Default: markdown) | | --output OUTPUT | Path for the generated output file (Default: API_Docs.md) | | --use-ollama | Enhance with a local Ollama model | | --ollama-model MODEL | Ollama model to use (e.g., phi3:mini) | | --use-openai | Enhance with an OpenAI model | | --openai-model MODEL | OpenAI model to use (e.g., gpt-4o-mini) | | --use-google-gemini | Enhance with a Google Gemini model | | --google-gemini-model MODEL | Gemini model to use (e.g., gemini-1.5-flash) | | --use-groq | Enhance with a Groq model | | --groq-model MODEL | Groq model to use (e.g., llama3-8b-8192) |

📝 Examples

Basic Generation:

# Generate Markdown docs from the current project
apimatic generate --src . --output API_Docs.md

AI-Enhanced Documentation:

First, set your key:

apimatic config --set-openai-key sk-xxxxxxxx

Then, generate with enhancement:

# Use OpenAI's gpt-4o-mini model
apimatic generate --src . --use-openai --openai-model gpt-4o-mini

# Use a local Ollama model
apimatic generate --src . --use-ollama --ollama-model phi3:mini

🤖 Recommended AI Models

| Provider | Recommended Model | Notes | | --- | --- | --- | | Ollama (Local) | phi3:mini, llama3:8b | Fast, free, and runs on your machine. Great for privacy. | | OpenAI | gpt-4o-mini | Excellent balance of cost, speed, and intelligence. | | Google Gemini | gemini-1.5-flash | Fast and cost-effective model from Google. | | Groq | llama3-8b-8192 | Incredibly fast inference speeds. |

⚙ Requirements

• Node.js (v16+ recommended)

• Python 3.9+ installed and available as python or python3

• Apimatic Python package installed:

  pip install Apimatic

🤝 Contributing

Contributions are welcome! Please fork the repo, make your changes, and submit a PR.

📄 License

This project is licensed under the MIT License.

✨ This version is tailored for npm users, with:

• npm badges instead of PyPI
• npm install -g Apimatic-cli instructions
• Keeps the same CLI usage table so both pip & npm users get consistent docs