A voice-first image editing skill that enables iterative generation, version branching, and multi-model management through a structured canvas system.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install babblebrush
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install babblebrush using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is babblebrush?
babblebrush is a professional-grade image editor designed for seamless integration with AI agents. It functions as a workspace where both humans and agents can generate, view, and manage images programmatically. Unlike static generation tools, babblebrush organizes work into canvases and versions, allowing for a non-linear editing process where users can branch off from any point in history to try new directions.
This skill is built for the Openclaw Skills ecosystem to provide high-level control over image creation. It supports a Bring Your Own Key (BYOK) model, enabling developers to use their own credentials for providers like Google Gemini and xAI. By using babblebrush, agents can perform complex tasks such as adding specific elements to a scene, changing artistic styles, or blending multiple reference images into a cohesive final product.
babblebrush Use Cases
- Creating unique brand assets or logos from scratch using text-to-image prompts.
- Performing iterative edits on existing photos to modify backgrounds or add specific objects.
- Branching a design project into multiple versions to compare different AI model outputs.
- Using multiple reference images to guide the AI in maintaining consistent visual context.
- Reverting a project to a previous state using version promotion for instant time travel.
How babblebrush Works
- The user or agent initializes a workspace by creating a new canvas or uploading an existing image to start from.
- An edit request is enqueued as a new canvas version, specifying a prompt and optionally a specific source version to branch from.
- If specialized context is needed, reference images are uploaded to the source version before the generation process begins.
- The skill processes the image generation asynchronously through the selected AI provider (e.g., Gemini or xAI).
- The agent polls the version status endpoint to monitor progress until the generation reaches a terminal state.
- Completed images can be promoted to the current version, shared via public URLs, or used as the basis for further iterative edits.
babblebrush Setup
To get started with this skill in the Openclaw Skills framework, you need to obtain an API key from the babblebrush settings page. Configure your environment by setting the following variable:
export BABBLEBRUSH_API_KEY="your_bb_token_here"
You can then verify your setup and check your available credits or platform access status using the following command:
curl -H "Authorization: Bearer $BABBLEBRUSH_API_KEY" "https://babblebrush.com/api/v1/me"
babblebrush Data Schema & Taxonomy
The skill organizes image data using a hierarchical structure to maintain history and context within Openclaw Skills:
| Object | Description |
|---|---|
| Canvas | The primary container for an image project, identified by a unique public token. |
| Canvas Version | An individual edit or state within a canvas, tracking prompts, models, and status. |
| Provider Model | Technical metadata for AI engines, including image limits per prompt. |
| Reference Images | Supplementary image data uploaded to a version to influence future edits. |
| Credentials | Securely stored API keys for external providers like Google or xAI. |
babblebrush Advanced Features
- Multi-agent support for collaborative image editing and management.
- Version branching to allow parallel creative paths without overwriting previous work.
- Time travel capabilities to promote any historical version to the current primary image.
- Flexible model overrides allowing users to switch AI providers (Gemini vs xAI) on a per-edit basis.
- Support for multiple reference images to provide deep visual context for complex blending tasks.
- Asynchronous job management with support for cancelling pending or processing edits.
SKILL.md
Loading
Related Openclaw Skills
x-alive
A comprehensive toolkit for launching and maintaining an authentic, human-like AI presence on X through organic engagement and real-time situational awareness.
notion-sync
A CLI tool and library for mirroring local directory structures and file content directly into Notion pages.
FRED Navigator
An advanced AI agent skill for navigating the Federal Reserve Economic Data (FRED) database using natural language and dual-layer validation.
SerpApi Flights (Google Flights)
A powerful flight search skill for querying real-time Google Flights data including prices, schedules, and aircraft details.
Soft Pillow
A specialized AI skill for retrieving and analyzing sleep logs, dream journals, and mood data from the Soft Pillow iOS application.
AllStar Link Node Control
A specialized client for monitoring and controlling AllStar Link amateur radio nodes through a secure REST API using Openclaw Skills.
Featured*
