SkillBoss for Openclaw

A comprehensive multi-AI gateway providing unified access to over 50 models for text, vision, audio, and search via a single API key.

yshuolu
v1.0.0
Mar 11, 2026
0
532
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install skillboss-run

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 skillboss-run 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?

SkillBoss is a powerful integration for developers who need to pivot between diverse AI capabilities without managing dozens of separate provider accounts. As a premier entry in the Openclaw Skills ecosystem, it consolidates access to industry-leading models from providers like Bedrock, Minimax, and specialized multimedia engines.

Whether you are generating high-fidelity images, synthesizing lifelike speech, or querying the latest LLMs for complex reasoning, SkillBoss acts as a standardized abstraction layer. This reduces friction in building multi-modal applications and ensures that your agentic workflows remain flexible and vendor-agnostic.

SkillBoss Use Cases

  • Rapidly switching between different LLM providers like Claude and local models to compare performance.
  • Automating media generation pipelines by piping text prompts into image, video, or music models.
  • Implementing text-to-speech (TTS) features into CLI tools or bots using a single environment variable.
  • Discovering and testing new AI models as they are released via the dynamic model listing feature.

How SkillBoss Works

  1. The user provides a valid SKILLBOSS_API_KEY which serves as the universal credential for all supported providers.
  2. Developers use the models command to fetch a real-time list of available endpoints and capabilities.
  3. A specific model or task type is invoked via the run.mjs script, passing the provider/model string and the desired prompt.
  4. SkillBoss processes the request through its gateway and returns the structured response or a direct URL for generated media.
  5. For media tasks, the output URL can be immediately consumed by downstream processes like curl for local storage.

SkillBoss Setup

To use this contribution to Openclaw Skills, ensure you have Node.js installed and set up your API key:

# Export your API key obtained from skillboss.co
export SKILLBOSS_API_KEY='your_api_key_here'

# List available models to verify connection
node scripts/run.mjs models

SkillBoss Data Schema & Taxonomy

SkillBoss utilizes a clean, command-driven data structure to manage multi-modal outputs:

Data Component Format Description
Model Registry JSON List Dynamic list of providers and model IDs returned by the models command.
Text Outputs Stdout Direct string responses for chat and research tasks.
Media Assets URL String Temporary public URLs generated for images, video, and audio files.
Environment Env Var Centralized authentication via SKILLBOSS_API_KEY.

SkillBoss Advanced Features

  • Task-based abstraction: Use the task command to let the gateway select the best model for a specific domain (e.g., image, chat, tts).
  • Cross-provider consistency: Standardized command syntax regardless of whether you are calling Bedrock, Minimax, or other vendors.
  • Automated media ingestion: Easily scriptable output for pipe-based workflows, such as automatically downloading generated assets via curl.
  • Category-specific discovery: Filter the model list by type (e.g., models image) to streamline tool selection for Openclaw Skills developers.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*