rey-model-router
Automatically select optimal model based on task complexity. Routes simple tasks to Sonnet 4.5 (cost-effective) and complex tasks to Opus 4.6 (high-quality). Always active.
Why use this skill?
Optimize AI costs with rey-model-router. Automatically route tasks between Claude Sonnet and Opus for perfect balance of performance and budget in OpenClaw.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/sa9saq/rey-model-routerWhat This Skill Does
The rey-model-router is an intelligent autonomous skill designed to optimize the performance-to-cost ratio of your OpenClaw AI agent. By dynamically evaluating user intent, it acts as a gatekeeper that routes incoming tasks to either Claude 3.5 Sonnet (for routine tasks) or Claude 3.5 Opus (for complex engineering or analytical requirements). It handles the overhead of model selection, ensuring high-quality results for complex code development while preventing unnecessary expenditure on simple queries.
Installation
To integrate this router into your workflow, use the OpenClaw command-line interface. Run the following installation command in your terminal:
clawhub install openclaw/skills/skills/sa9saq/rey-model-router
Once installed, the router operates as a background process, perpetually monitoring your request stream for specific trigger keywords and complexity markers.
Use Cases
- Software Development: Automatically scales up to Opus when the agent detects needs for architectural design or complex debugging.
- Content Creation: Keeps routine social media updates and summary tasks on the cost-efficient Sonnet model.
- Hybrid Workflows: Ideal for teams managing a mix of administrative tasks and high-stakes technical projects, ensuring you get the right precision exactly when you need it.
- Automated Fallback: The skill includes specialized logic to toggle between subscription-based model access and API-based usage, ensuring service continuity even if rate limits are reached.
Example Prompts
- "Draft a LinkedIn post announcing our new product launch based on the provided press release."
- "I need to debug this Google Apps Script; it keeps throwing a 403 error during the OAuth process."
- "Summarize the last three weeks of Slack logs and highlight the top 5 technical blockers identified by the team."
Tips & Limitations
- The 2-Point Rule: If you are unsure of the complexity, use the self-evaluation checklist. If two or more criteria (such as >50 lines of code, system design, or multi-step reasoning) are met, the agent will prioritize the Opus model.
- Manual Overrides: The automatic router is a heuristic tool. If you require a specific model for a task, you can always bypass the logic using the
/modelcommand. - Monitoring: Always check the post-execution footer (e.g., "💰 Sonnetで処理しました") to verify which model was utilized for your specific task.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-sa9saq-rey-model-router": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: external-api
Related Skills
threat-model
Threat modeling and attack scenario design. Identify risks before they become vulnerabilities. STRIDE, attack trees, risk matrix.
Sns Auto Poster
Schedule and automate social media posts to X/Twitter with cron-based queue management.
security-review
Comprehensive security review for code, configs, and operations. OWASP, prompt injection, crypto security. Auto-triggers on security-related changes.
Process Monitor
Monitor system processes, identify top CPU/memory consumers, and alert on resource thresholds.
Readme Generator
Auto-generate comprehensive README.md files by analyzing project structure and configuration.