ask-my-tti

Zero-dependency AI image generation module. Works with any OpenAI-compatible image API (DALL-E, etc.).

Install

npm install ask-my-tti

Configure

npx ask-my-tti config

Reuses the same ~/.askttirc config. Prompts for Base API URL, API Key, and Model (default: dall-e-3).

CLI usage

npx ask-my-tti draw "Cow" cow.jpg

Library usage

const amtti = require("ask-my-tti");

// Sync
amtti.drawSync("Cow", "cow.jpg");

// Async
await amtti.draw("Cow with a calf", "cows.jpg");

// With options
amtti.drawSync("Cow", "widecow.jpg", { size: "1792x1024" });
amtti.drawSync("Cow", "lowcow.jpg", { quality: "low" });

// Get current model
console.log(amtti.getCurrentModel());

Options

All options are optional:

| Option | Default | Description | |--------|---------|-------------| | size | 1024x1024 | Image size (e.g., 1792x1024, 1024x1792) | | quality | auto | Generation quality (auto, hd, standard, low) | | style | vivid | Image style (vivid, natural, auto) | | model | from config | Override model name |

The output format is inferred from the file extension (.jpg → jpeg, .png → png, .webp → webp).

API

amtti.drawSync(prompt, outputPath, options?): { outputPath, revisedPrompt }

Generates an image synchronously and saves to outputPath. Returns metadata including revisedPrompt if the API rewrote the prompt.

amtti.draw(prompt, outputPath, options?): Promise<{ outputPath, revisedPrompt }>

Async version.

amtti.getCurrentModel(): string

Returns the model from the active config.

License

Unlicense — see LICENSE.