Back to directory

Hypit

A domain-specific language and toolkit that lets an AI coding agent clone a video's structure, such as its footage, captions, B-roll and effects anchored to words rather than timestamps, and produce multiple variants from one command.

Hypit

Hypit gives AI agents such as Claude Code and Codex a language and system for building videos. Given a reference video, an agent can clone it as a complete workflow, reconstructing its footage, captions, B-roll and effects, anchored to words instead of seconds.

Cloning an existing video is one way to start; a workflow can also begin from a template or from a plain-language description that the agent turns into a workflow from scratch. Generation models are optional: a workflow can compile captions, motion graphics and code-rendered visuals into a finished video without calling an image or video generation model.

It installs as an agent skill through a package manager command, and works with a recommended hosted model service or with a user's own API or local models. The core tool is free to use, though the coding agent and any model services used have their own accounts and charges.

More in Generative media

See all 146
  • Blender Copilot

    github.com

    A Copilot-style AI chat panel inside Blender that runs its agent loop in Blender's own Python process, executes the model's generated code against the live scene, and wraps every conversational turn in a single undo step.

  • A ComfyUI add-on that patches ComfyUI-GGUF so Qwen-Image-2.1 models can run fully as GGUF files, fixing a text-encoder mismatch and missing architecture detection until upstream support arrives.

  • A ComfyUI node that expands a short prompt into a full styled prompt, negative prompt and aspect ratio for Qwen-Image-2.1, using one of 17 photographic styles.