A unified API integration providing access to over 50 AI models for text, image, video, search, and communication tasks.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install obsidians
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 obsidians 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 SkillBoss AI Multi-Model Integration?
SkillBoss acts as a powerful orchestration layer for AI development, offering a single point of access to top-tier providers including OpenAI, Anthropic, Google Vertex, ElevenLabs, and Replicate. By implementing this within Openclaw Skills, developers can eliminate the friction of managing multiple API keys and complex SDKs, instead using a standardized interface to call upon specialized models for any technical requirement.
Whether you need state-of-the-art language models like Claude 3.5 Sonnet, advanced image generation via Flux, or specialized tools for document parsing and web scraping, this skill provides the infrastructure to build versatile agents. The platform handles the heavy lifting of routing and authentication, allowing you to focus on building features rather than managing infrastructure.
SkillBoss AI Multi-Model Integration Use Cases
- Generating high-fidelity images and videos for marketing or design directly from text prompts.
- Automating document workflows by parsing PDFs and DOCX files into clean Markdown or structured JSON.
- Implementing multi-channel communication including transactional emails and SMS verification codes.
- Conducting deep web research and scraping using specialized search models like Perplexity and Firecrawl.
- Building accessibility features with high-quality text-to-speech and accurate speech-to-text transcription.
How SkillBoss AI Multi-Model Integration Works
- Authentication is handled via a single SKILLBOSS_API_KEY environment variable used for all 50+ supported models.
- Developers can browse the live models endpoint to discover available capabilities, model IDs, and specific pricing or performance tiers.
- Requests are sent to a unified base URL where the skill routes the payload to the specific provider (e.g., Bedrock, Vertex, or Replicate).
- For complex workflows, the Pilot mode can be utilized to automatically select the most cost-effective or highest-quality model for a specific task type.
- The skill returns standardized responses, making it easy to pipe output from one model (like a chat completion) into another (like image generation).
SkillBoss AI Multi-Model Integration Setup
To get started, you must obtain an API key from the SkillBoss dashboard and configure your environment.
# Export your API key to your environment
export SKILLBOSS_API_KEY="your_key_here"
# Test the integration by listing available models
curl -s https://api.heybossai.com/v1/models \
-H "Authorization: Bearer $SKILLBOSS_API_KEY"
SkillBoss AI Multi-Model Integration Data Schema & Taxonomy
The skill utilizes a unified request and response structure across different model categories to simplify data handling within Openclaw Skills.
| Category | Request Parameter | Primary Output |
|---|---|---|
| Chat | messages, model |
choices[0].message.content |
| Image/Video | prompt, inputs |
image_url or video_url |
| Audio | text or audio_data |
audio_url or text |
| Search | query, depth |
searchResults or sourcedAnswer |
| Documents | document_url |
Markdown or structured schema |
SkillBoss AI Multi-Model Integration Advanced Features
- Smart Pilot Mode: An intelligent routing system that auto-selects the best model for a given task (image, text, search) to optimize cost and latency.
- Multi-Provider Failover: Access redundant models across different clouds (AWS Bedrock, Google Vertex, etc.) to ensure high availability.
- Image-to-Video Workflow: Support for advanced generative tasks where an existing image can be used as a seed for video motion.
- Structured Data Extraction: Powerful document processing capabilities that convert unstructured files into specific JSON schemas using the Reducto engine.
SKILL.md
Loading
Related Openclaw Skills
Nano Banana Pro (SkillBoss)
A unified API interface providing access to over 50 premier AI models for text, image, video, and utility tasks through a single endpoint.
Nanobanana Image Generation
A powerful AI skill for generating professional images and exact scientific plots using Google Gemini models and Python plotting tools.
SkillBoss: Multi-Model AI Orchestrator
SkillBoss provides a single API gateway to over 50 leading AI models across providers like OpenAI, Anthropic, and Google for chat, media, and search.
SkillBoss AI Multi-Model Gateway
A unified API gateway providing seamless access to over 50 AI models across multiple providers including OpenAI, Anthropic, and Google.
SkillBoss AI Multi-Model Suite
A unified API gateway providing access to 50+ multimodal AI models and productivity tools with smart routing and a single authentication key.
nano-pdf: Multi-Model AI Powerhouse
A versatile AI skill providing a single interface for 50+ state-of-the-art models covering text, vision, audio, and document processing.
Featured*
