An intelligent model routing system that automatically selects the best LLM for any task to reduce costs by 60%.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-router
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 openclaw-router 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 OpenClaw Router?
OpenClaw Router is an advanced orchestration layer designed to optimize LLM usage by intelligently routing requests between local and cloud-based models. By analyzing task requirements across five dimensions—knowledge, reasoning, context, quality, and tools—it ensures that simple queries are handled by free local models like Ollama, while complex reasoning is escalated to top-tier providers like OpenAI or Anthropic.
This approach is essential for developers using Openclaw Skills who want to maintain high performance while significantly reducing operational overhead. The system supports local-first, cloud-only, or hybrid deployment strategies, making it a versatile choice for both personal productivity and enterprise-scale AI integration.
OpenClaw Router Use Cases
- Automatically routing simple coding questions to local Qwen models to save API credits.
- Escalating complex strategic analysis or creative writing tasks to GPT-4 or Claude-3 for higher quality results.
- Managing monthly AI budgets with real-time token tracking and automated cost alerts.
- Improving data privacy by prioritizing local model execution for sensitive internal information.
- Optimizing response latency by selecting models based on task category and required reasoning depth.
How OpenClaw Router Works
- The skill intercepts an incoming task and performs a 5-dimensional assessment of the required capabilities (knowledge, reasoning, context, quality, and tools).
- It identifies the specific task category, such as code generation, creative writing, or data analysis.
- The router checks the user's predefined preferences and current budget status stored in the configuration file.
- It selects the most cost-effective model (Local first, then L2 Cloud, then L3 Expert) that meets the complexity threshold.
- The request is executed through the chosen provider, and the system tracks token usage and provides real-time cost reporting.
OpenClaw Router Setup
To integrate this routing capability into your environment, use the following commands:
# Install via ClawHub
clawhub install openclaw-router
# Initialize the configuration wizard to set up your model pool
openclaw router config --init
# Enable the routing service to start managing requests
openclaw router enable
OpenClaw Router Data Schema & Taxonomy
The skill manages its configuration and logs within the user's local environment to ensure privacy and control.
| File/Component | Description |
|---|---|
~/.openclaw/router_config.yaml |
The primary configuration file containing model definitions, threshold settings, and budget limits. |
Thresholds |
Logic defined as 'balanced', 'performance', or 'cost' to determine when to escalate tasks. |
Budget Schema |
Tracks monthly spending in multiple currencies (USD, CNY, EUR) with alert triggers. |
Model Pool |
A list of local (Ollama) and cloud (DashScope, OpenAI, Anthropic) endpoints. |
OpenClaw Router Advanced Features
- Intelligent preference learning that adapts model selection based on historical usage patterns.
- Multi-cloud and hybrid deployment support for maximum availability and cost efficiency.
- Granular budget management with multi-stage alerts at 50%, 80%, and 95% of usage.
- Enterprise-grade features including multi-user management, custom model pools, and API access.
- Real-time token tracking and transparent cost reporting for every request processed through Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Gmail Manager
An intelligent Gmail management assistant that automates inbox triage, email organization, and productivity workflows via the Rube MCP.
Mista Sovereign Agent
An autonomous digital agent for Moltbook that utilizes neural resonance and predatory skill assimilation for digital dominance.
Mista Sovereign Agent
Mista Sovereign is an autonomous digital agent for Moltbook designed for intelligence extraction, neural resonance, and predatory skill assimilation.
ElevenLabs Agents Management
Create, manage, and deploy ElevenLabs conversational AI agents seamlessly using automated CLI workflows.
OpenKM REST Document Management
A Python-based CLI skill for automating OpenKM document management, metadata organization, and workflow execution via REST API.
Moral Compass
An ethical reasoning framework for AI agents grounded in universal principles like justice and truthfulness.
Featured*
