Dreamer LLM for Openclaw

A unified API gateway to access 12+ AI model providers for chat, vision, and media generation through a single, streamlined interface.

lukeslp
v1.0.0
Feb 18, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install dreamer-llm

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install dreamer-llm 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 Dreamer LLM?

Dreamer LLM is a versatile integration designed for Openclaw Skills that provides a single interface to interact with the world's leading AI models. Instead of managing dozens of individual API keys and navigating disparate SDKs for Anthropic, OpenAI, xAI, and others, developers can leverage a streamlined endpoint at api.dr.eamer.dev. This skill simplifies the process of building multi-model applications, enabling seamless provider switching and comparative analysis to find the best model for any specific task within your Openclaw Skills workflow.

By unifying access to providers like Mistral, Cohere, and Perplexity, Dreamer LLM reduces the architectural overhead for developers. Whether you are generating images, converting text to speech, or conducting complex vision analysis, this skill provides a consistent framework for all your generative AI needs.

Dreamer LLM Use Cases

  • Comparing output quality and performance across multiple LLM providers using Openclaw Skills.
  • Accessing a wide array of premium AI models without the need for individual direct provider accounts.
  • Rapidly prototyping multi-modal features including vision analysis, image generation, and text-to-speech.
  • Integrating high-performance models from Groq or specialized research tools from Perplexity into Openclaw Skills workflows.

How Dreamer LLM Works

  1. Configure the DREAMER_API_KEY environment variable to authenticate with the unified Dreamer gateway.
  2. Define the desired provider (e.g., anthropic, openai) and model in your JSON request payload.
  3. Send the request to the central Dreamer API endpoint to route it to the appropriate model host.
  4. Receive the processed response for chat, vision, or media generation via the Openclaw Skills interface.

Dreamer LLM Setup

To get started with this skill, obtain an API key from dr.eamer.dev. Once you have your key, configure your environment using the following command:

export DREAMER_API_KEY=your_key_here

Ensure this variable is accessible to your environment to enable authenticated requests. You can verify your setup by calling the models endpoint to see all available providers supported by Openclaw Skills.

Dreamer LLM Data Schema & Taxonomy

Dreamer LLM uses a standardized JSON schema for all interactions, making it easy to swap models without changing your core logic.

Parameter Type Description
provider String The model host (e.g., anthropic, openai, xai, groq, ollama)
model String The specific model identifier (e.g., claude-sonnet-4-5-20250929)
messages Array A list of chat message objects containing roles and content
image_url String The URL of the image to be processed during vision analysis
text String The input string used for text-to-speech generation

Dreamer LLM Advanced Features

  • Unified access to 12 distinct model providers including HuggingFace, Ollama, and Google Gemini.
  • Cross-provider vision analysis allowing you to describe and analyze images using top-tier vision models.
  • Integrated Image Generation and Text-to-Speech (TTS) endpoints for full multi-media support within Openclaw Skills.
  • Dynamic model discovery via the GET models endpoint to stay updated with the latest releases from all providers.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*