A high-performance AI image generation tool powered by the Coze Seedream 4.5 workflow for seamless visual creation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install coze-image-2
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 coze-image-2 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 Coze Image Generation?
Coze-Image is a specialized integration designed for Openclaw Skills that leverages the Seedream 4.5 model via the Coze API. This skill enables developers and AI agents to transform textual descriptions into high-quality visual assets by executing pre-defined workflows on the Coze platform. By acting as a bridge between agent logic and advanced creative engines, it streamlines the process of generating AI-driven art directly from a command-line interface or agent interaction.
The tool is specifically optimized for users who require professional-grade image synthesis within their automated workflows. As part of the broader ecosystem of Openclaw Skills, it provides a structured way to handle API tokens, workflow IDs, and prompt processing to ensure reliable and repeatable visual outputs.
Coze Image Generation Use Cases
- Generating realistic or stylized images from natural language text prompts.
- Automating creative asset creation for digital marketing and content production.
- Providing AI agents with visual output capabilities during interactive conversations.
- Integrating professional AI drawing workflows into terminal-based developer environments.
How Coze Image Generation Works
- The user triggers the skill by providing a descriptive text prompt for the desired image.
- The skill validates the configuration and sends an authenticated request to the Coze API workflow endpoint.
- The Coze platform processes the request using the Seedream 4.5 engine based on the specified workflow ID.
- The system captures the resulting image URL from the API response and returns it to the user or agent.
Coze Image Generation Setup
To use this integration within Openclaw Skills, follow these configuration steps:
- Log in to coze.cn and navigate to API Management to create a personal API Token.
- Create or identify an image generation workflow to obtain your Workflow ID from the URL.
- Update the
CONFIGobject incoze_image.pywith your credentials:
CONFIG = {
"api_token": "YOUR_COZE_API_TOKEN",
"workflow_id": "YOUR_WORKFLOW_ID",
"api_url": "https://api.coze.cn/v1/workflow/run"
}
- Execute the generation command via the CLI:
python coze_image.py generate "your image description here"
Coze Image Generation Data Schema & Taxonomy
The skill manages data through API interaction and local configuration variables as follows:
| Attribute | Description | Source |
|---|---|---|
| api_token | Authentication secret for Coze cloud services | User Config |
| workflow_id | Unique identifier for the Seedream 4.5 workflow | User Config |
| prompt | The descriptive text input used for image synthesis | User Input |
| output_url | The generated public image link (e.g., https://s.coze.cn/t/...) | API Response |
Coze Image Generation Advanced Features
- Support for custom Coze workflows, allowing users to swap Seedream 4.5 for other specialized visual models.
- Direct integration with multi-agent pipelines for automated content generation within Openclaw Skills.
- Real-time execution and URL retrieval for high-speed feedback loops in dev environments.
- Cross-platform compatibility across macOS, Linux, and Windows systems.
SKILL.md
Loading
Related Openclaw Skills
Task Manager Skill
A professional task tracking and reporting system for AI agents that maintains a structured Markdown Kanban board and daily activity logs.
Google Search Web Skill
A powerful Python-based skill for AI agents to fetch real-time Google search results including titles, links, and summaries.
Hyperframes Video Studio
Hyperframes Video Studio is a guided OpenClaw Skill for creating deterministic, professional videos using reusable HTML templates and local-first rendering.
Qwen Web Agent
A robust Playwright-based automation tool for interacting with the Qwen web interface, supporting persistent sessions and streaming response detection.
BrainASLab FreeSurfer Asymmetry Analyzer
BrainASLab is an automated neuroimaging tool designed to calculate cortical asymmetry indices and generate publication-ready visualizations from FreeSurfer statistics.
HarmonyOS FlyVerify Integration
A specialized technical guide for integrating MobTech FlyVerify one-click login and verification services into HarmonyOS NEXT applications.
Featured*
