OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Dramaclaw

Dramaclaw is a self-hostable AI video production application that combines a node-based canvas for working with visual and audio assets with a pipeline for turning manuscripts or screenplays into finished films. Its built-in Xia Director agent can answer project questions, drive pipeline tasks and work with canvas node graphs.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Dramaclaw — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Dramaclaw when you want a self-hosted production workflow that connects story development, storyboard and asset creation, voice-over and final video assembly, or a node-based canvas for generative media work.

Who it is for

  • Creators developing AI-assisted visual stories
  • Indie studios producing dramas or serialized video
  • Creative engineers building generative media workflows
  • Teams making advertising, product videos or otome-style game content

Documented uses

  • Structure a manuscript and plan episodes
  • Write scripts and create storyboards and first frames
  • Generate and connect images, video, audio, 3D sets and scripts on a node-based canvas
  • Synthesize voice-over and assemble a finished film
  • Use an agent to answer project questions and drive pipeline or canvas tasks
Before you choose

Requirements & limits

The repository documents Docker deployment and a standalone Elastic License 2.0 offering. It says commercial use of the standalone version is permitted with a small “Powered by DramaClaw” notice, while packaging it as a hosted SaaS for others is not permitted under the stated license.

  • Inference runs through an OpenAI-compatible gateway; model inference is not performed on the host machine.
  • The documented gateway choices are the official RelayClaw service or a bundled gateway configured with upstream provider channels.
  • Docker and docker compose are the recommended deployment prerequisites; the README recommends 2 vCPU and 4 GB RAM.
  • A GPU is not required for the standard pipeline; the optional voxel or panorama-to-3D world extra requires a GPU and CUDA image.
  • The standalone commercial-use terms differ from the restriction on offering Dramaclaw as a hosted SaaS.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start DramaClaw with Docker Compose

Build and start the DramaClaw services from source using Docker Compose.

Before you start

  • Docker and Docker Compose.
  • Outbound network access to the official gateway or model providers configured in the bundled gateway.
  • For source builds, both DramaClaw and the bundled dramaclaw-gateway repositories cloned side by side.

Steps

  1. Clone the DramaClaw repository and the dramaclaw-gateway repository side by side.

  2. Change to the dramaclaw directory and copy .env.example to .env.

  3. Edit .env to set PROMPT_EXPORT_PASSWORD to a non-default value.

  4. Run the command below to build and start the API, bundled gateway, and web services.

    Documented command
    docker compose up -d --build

Expected outcome

The three services are started; the documented web UI is at http://localhost:8080 and the REST API is at http://localhost:8780.

Guide limitations

  • The source build depends on Docker Compose and the bundled gateway checkout.
  • Model inference runs through an OpenAI-compatible gateway rather than on the local machine.
  • The optional world extra for voxel or panorama-to-3D requires a GPU and CUDA image.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
dramaclaw/dramaclaw · Official homepage title · retained in its source language
How we review sourcesSuggest a correction