Abracadabrax Images & Video

Create from text prompts

Create images and short videos from text in your Abracadabrax workspace. Prepare a project in the conversation, choose a supported model, aspect ratio and video length, then review the settings before requesting generation.

What you can ask

Create a product image of a ceramic mug in morning light

MCP connector

Server URL: https://abracadabrax.app/api/chatgpt/mcp
Login: OAuth
Protocol: MCP 2025-11-25
Instructions: Call first (tab image or video, prompt, optional model_id/ratio/duration). If the user explicitly asked to generate, call with the prepared project.id; it spends credits. Otherwise show the settings and wait for explicit confirmation. The widget Generate button is optional: complete requested generations through the tools on clients without a widget. Use to check progress and return the output. Without a linked account the tools still work: then returns state account_required with dashboard_url; tell the user to open Abracadabrax from the widget or that link, where they sign in or create the account and finish this generation, and do not ask them to link the app first. When completed, give the user the absolute media URL from the tool result as a plain link (ChatGPT does not render external images inline; the widget shows the media) so they can open or download it; never illustrate the result with web or image search. If the user asks to reopen a prepared project, call with the project.id. Use this only for text-to-image or text-to-video prompts. Do not use it for uploaded, attached or reference images, image editing, background removal, upscaling, virtual try-on, image-to-video or other media-to-media workflows. If the user asks to use an uploaded, attached or reference image or video, do not call these tools; explain that the app only creates images and videos from text and offer to open Abracadabrax for upload-based workflows.

Tools

Changes data1

Deletes or can't be undone4

Skills

None in the store listings or on the server

MCP connector

Server URL

https://abracadabrax.app/api/chatgpt/mcp

OAuth

Protocol

MCP 2025-11-25

Instructions

Call first (tab image or video, prompt, optional model_id/ratio/duration). If the user explicitly asked to generate, call with the prepared project.id; it spends credits. Otherwise show the settings and wait for explicit confirmation. The widget Generate button is optional: complete requested generations through the tools on clients without a widget. Use to check progress and return the output. Without a linked account the tools still work: then returns state account_required with dashboard_url; tell the user to open Abracadabrax from the widget or that link, where they sign in or create the account and finish this generation, and do not ask them to link the app first. When completed, give the user the absolute media URL from the tool result as a plain link (ChatGPT does not render external images inline; the widget shows the media) so they can open or download it; never illustrate the result with web or image search. If the user asks to reopen a prepared project, call with the project.id. Use this only for text-to-image or text-to-video prompts. Do not use it for uploaded, attached or reference images, image editing, background removal, upscaling, virtual try-on, image-to-video or other media-to-media workflows. If the user asks to use an uploaded, attached or reference image or video, do not call these tools; explain that the app only creates images and videos from text and offer to open Abracadabrax for upload-based workflows.