Skip to content
Open studio
An AI teammate that keeps learning from experience

Your AI
production assistant.

Describe the shot. Your assistant handles production. Review the take, give notes, and continue to the next version.

Open studioBuilt for screenwriters and AI filmmaking teams
Illustrative scene: a white cup in a rain-washed courtyardIllustrative image
Morning courtyardShot 02
Example workflow

Keep the courtyard and composition. Make the steam more subtle.

  1. Read the briefClarify the shot
  2. Create a takeUse your platform
  3. Your reviewChoose what is next

From a shot brief to a take you choose.

  1. 01

    Describe the shot

    Explain the image you have in mind and the details that must stay consistent. Your assistant starts with the same brief.

    “A courtyard after rain. Push in slowly. Keep the soft morning light.”
  2. 02

    Let it produce

    Your assistant works through a connected platform. Find progress, candidate takes, and the original canvas in one studio.

    Shot brief → Platform canvas → Generated take
  3. 03

    Review and refine

    Choose a take, or say what to keep and what to change. Your feedback follows the next version.

    “Keep the framing. Make the steam more subtle.”

Change platforms.
Keep the experience.

The same assistant carries its memory and working methods to each connected platform, with the creative context you have already shared.

Shot versions, original takes, and review notes stay connected. Each change has a record; each new take has context.

Carry your notes forwardExample
Keep
Courtyard, framing, soft morning light
Change
Lighter steam and a slower push-in
Preserve the original and notes. Create a new take.

Work with your production platforms.

Your assistant uses your authorized account in the platform’s own creative environment. Supported workflows will expand over time.

LibTV

Create a canvas, generate a single shot, and retrieve the result.

Connected
Xiaoyunque

Live account connection is not available yet.

Coming later

LibTV currently supports Wan 2.2 text-to-video: 5 seconds, 720P, 16:9, one take. You confirm each generation before submission.

Before your first shot.

Who decides whether a take works?

You do. The assistant handles production and gathers results. You review the take and choose to accept it, use part of it, or request another version.

How is this different from a platform’s built-in agent?

Chouka Assistant connects the same Hermes assistant to different platforms. Hermes manages its native memory and skills, while shot versions and review records stay in your production workspace.

Does video generation cost money?

It may. Charges come from the platform you use. Chouka Assistant shows the requested settings and asks for confirmation before submission. LibTV currently does not return a reliable price quote; the confirmation screen makes this clear.

Can I buy a membership now?

Monthly memberships are provisioned for the pilot by an administrator. Payments are not open yet. Assistant usage and production platform charges are tracked separately.

Make the next shot together.

Start with one shot. See what your assistant produces, then tell it how to shape the next version.

Open studio