XRYPTON AI Commands — Ask, Imagine, and Text-to-Speech

Use XRYPTON’s built-in AI to chat with a language model, generate images from text prompts, and convert text to spoken audio files.

XRYPTON ships three AI-powered commands that let you interact with a language model, produce images from a written prompt, and synthesize speech — all without ever leaving Discord. Each command works as a standard prefix command (e.g. ,ask) and can take a moment to respond while the AI processes your request; XRYPTON will show a typing indicator so you know it is working.

Commands

Command	Usage	Description
ask	,ask <question>	Send a question or prompt to the AI and receive a text reply. Alias: gpt
imagine	,imagine <prompt>	Generate up to four images from a text description. Alias: prompt
tts	,tts <text> [voice]	Convert text to a speech audio file and post it in the channel.

---

ask

Aliases: gpt

Ask the AI any question or give it any instruction. XRYPTON sends your message to the language model and posts the response in an embed. If the reply exceeds 4 000 characters the full text is attached as a .txt file instead.

,ask What is the difference between a compiler and an interpreter?
,gpt Write me a haiku about rain.

Your prompt must be 4 000 characters or fewer. Prompts longer than that are rejected before the AI is ever contacted.

---

imagine

Aliases: prompt

Describe an image in plain English and XRYPTON will generate it for you. The bot returns up to four images, each displayed in its own embed with your original prompt shown as context.

,imagine a cyberpunk city at night with neon lights reflecting on wet pavement
,prompt a golden retriever wearing a wizard hat, oil painting style

Your image prompt must be 1 000 characters or fewer.

Be specific about style, lighting, and mood to get the best results — for example, mentioning “watercolour”, “cinematic”, or “pixel art” steers the model toward that aesthetic.

---

tts

Turn any text into a spoken-word audio file. XRYPTON synthesises the speech and posts the resulting .mp3 or .wav attachment directly in the channel. You can optionally choose a voice model by appending its name as the last word of your message.

,tts Hello, welcome to the server!
,tts Good morning everyone. aura

Optional voice parameter

Append a supported voice name as the final word of your text to switch voices. If no voice is specified the default aura voice is used.

,tts This is a test message luna

The text portion of your message must be 2 000 characters or fewer after the optional voice name is stripped. Providing only a voice name with no text will return an error.

Audio generation depends on an external TTS provider. If the provider is unavailable you will receive an error message. Try again after a short wait.