A powerful media generation skill leveraging Cloudflare Workers AI for professional text-to-image and text-to-speech capabilities.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cloudflare-media
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 cloudflare-media 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 Cloudflare Media?
Cloudflare Media is a sophisticated integration for AI agents that provides direct access to the Cloudflare Workers AI ecosystem. It enables developers and users to generate high-fidelity visual and auditory content through a curated selection of state-of-the-art models. This skill is a core component of the Openclaw Skills ecosystem, designed to streamline media production workflows without requiring complex local hardware.
The skill supports a diverse array of models, ranging from the high-performance FLUX.2 family and Stable Diffusion XL for imagery to advanced Deepgram Aura and MyShell MeloTTS for natural language synthesis. By centralizing these capabilities into a single interface, it allows for rapid prototyping and deployment of AI-driven creative tools.
Cloudflare Media Use Cases
- Automated generation of marketing visuals and social media assets using text-to-image prompts.
- Real-time voice synthesis for AI assistants, chatbots, and accessibility tools via high-quality TTS.
- Batch processing of image generation tasks using fast-inference models like FLUX.1 schnell.
- Creating localized audio content in multiple languages including English, Chinese, Japanese, and Korean.
How Cloudflare Media Works
- The agent identifies a media generation intent such as generate image or text-to-speech.
- Credentials (Account ID and API Token) are retrieved from local configuration files or requested from the user.
- The user is presented with a categorized list of available models and pricing for selection.
- The agent constructs a request with the specified prompt and parameters (e.g., dimensions, speaker style).
- Requests are dispatched to the Cloudflare Workers AI API via multipart/form-data or JSON body.
- The resulting binary stream or base64 data is processed and saved as .png, .jpg, .mp3, or .wav files.
Cloudflare Media Setup
To integrate this skill into your environment, follow these steps:
- Ensure you have a Cloudflare Account ID and an API Token with Workers AI permissions.
- Create a configuration file at
skills/cloudflare-media/config.jsonor update yourMEMORY.mdwith your credentials:
{
"CLOUDFLARE_ACCOUNT_ID": "your_account_id",
"CLOUDFLARE_API_TOKEN": "your_api_token"
}
- Deploy the Openclaw Skills package and ensure the cloudflare-media skill is enabled in your agent configuration.
Cloudflare Media Data Schema & Taxonomy
The skill manages its outputs and configurations through a structured directory system:
| File/Folder | Description |
|---|---|
skills/cloudflare-media/config.json |
Stores sensitive API credentials and default model preferences. |
*.png / *.jpg |
Output files for image generation tasks (varies by model). |
*.mp3 / *.wav |
Output files for speech synthesis (Aura-2 uses MP3, MeloTTS uses WAV). |
MEMORY.md |
Persistent memory storage for session-based user preferences. |
Cloudflare Media Advanced Features
- Support for Black Forest Labs FLUX.2 (4B, 9B, and dev variants) for varying levels of distillation and quality.
- Multi-language TTS support via MeloTTS covering Chinese, Japanese, Korean, French, Spanish, and English.
- Context-aware speech synthesis using Deepgram Aura-2 with 40 distinct speaker personas.
- Low-latency generation options using SDXL-Lightning and DreamShaper 8 LCM for real-time applications.
SKILL.md
Loading
Related Openclaw Skills
Spydra
Spydra is a high-performance web scraping and data extraction tool designed for researchers and businesses to programmatically collect web information.
Mambo TTS
A high-performance text-to-speech skill for Openclaw providing lively, pre-configured Chinese voice profiles based on Edge TTS.
Skill Creator
Skill Creator is a specialized meta-skill designed to help developers build, test, and refine new AI agent capabilities through an iterative, data-driven workflow.
Workspace Local Retrieval
A professional framework for building local-first retrieval architectures with strict security boundaries and automated maintenance workflows.
AI Earnings Tracker
An automated financial monitoring tool that tracks tech and AI earnings reports, generates AI summaries, and sets alert notifications.
1688-shopkeeper
A comprehensive e-commerce automation assistant for sourcing 1688 products and managing multi-platform store listings.
Featured*
