An intelligent routing system that automatically selects the most efficient Qwen model based on task complexity to optimize performance and cost.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install smart-model-selector
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 smart-model-selector 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 Smart Model Selector?
The Smart Model Selector is a powerful orchestration tool designed to enhance the efficiency of your AI workflows. By acting as an intelligent router, it analyzes the nature of every incoming request and dispatches it to the most suitable model in the Qwen family, such as qwen3.5-plus for simple tasks, qwen-max for complex reasoning, or qwen-coder-plus for specialized programming needs.
As a vital addition to your collection of Openclaw Skills, this tool doesn't just route—it learns. It tracks interaction success rates and user feedback to continuously refine its selection logic. This ensures that you receive high-quality results while significantly reducing unnecessary API costs, making your AI agent smarter and more economical with every conversation.
Smart Model Selector Use Cases
- Routing programming and debugging queries specifically to qwen-coder-plus for superior code generation.
- Using qwen-max for deep architectural planning, multi-step reasoning, and complex document analysis.
- Assigning lightweight queries like simple translations or general questions to qwen3.5-plus to minimize latency and cost.
- Auditing AI usage efficiency through detailed model performance statistics and success ratios.
How Smart Model Selector Works
- The skill intercepts the user's input and performs a semantic analysis to identify keywords and intent.
- It matches the intent against a logic matrix (e.g., code, long-form text, or simple Q&A).
- The system selects the optimal model and provides a transparent explanation of why that specific model was chosen.
- Interaction metrics, such as the number of turns to reach a solution, are recorded locally.
- Future routing decisions are optimized based on historical performance data and explicit user ratings.
Smart Model Selector Setup
Install the skill directly via the command line:
# Install via clawhub
clawhub install smart-model-selector
# Or manual installation
git clone https://github.com/yourusername/smart-model-selector.git
cp -r smart-model-selector ~/.openclaw/workspace/skills/
# Enable the skill
openclaw skills enable smart-model-selector
Smart Model Selector Data Schema & Taxonomy
The skill manages all data locally to ensure privacy and fast access:
| Data Component | Format | Description |
|---|---|---|
model_selection.db |
SQLite | Local database storing interaction history, model scores, and feedback logs. |
config.json |
JSON | Configuration file for setting default models and custom keyword rules. |
stats |
Metadata | Aggregated usage data displayed via the statistics command. |
Smart Model Selector Advanced Features
- Custom routing rules that allow users to map specific keywords (e.g., 'translate', 'creative') to preferred models.
- Manual model override capability using the
/model-usecommand for total control. - Detailed performance analytics via
/model-statsto track ROI and model reliability. - Feedback-driven learning where user ratings directly influence the weighting of future model selections.
- Seamless integration with existing Openclaw Skills to provide a unified, intelligent agent experience.
SKILL.md
Loading
Related Openclaw Skills
Climbing Gym Operations
An autonomous operations assistant for independent climbing gyms to manage memberships, route-setting schedules, and safety compliance.
X Trends
A specialized skill for retrieving real-time trending topics and hashtags from X (Twitter) based on specific geographic locations.
Weather Report Skills
A specialized skill for AI agents to fetch, process, and present localized weather information using highly structured Markdown templates.
Corporate Car Service Assistant
A professional AI-driven car service assistant designed for corporate travel, immediate ride-hailing, and airport logistics management.
Skill Security Checker
A comprehensive security auditor designed to scan and verify the safety of third-party AI agent skills before deployment.
PDF Label Data Extractor
An intelligent PDF data extraction tool that transforms complex product labels into structured Excel spreadsheets with high precision.
Featured*
