smart-model-switching
Auto-route tasks to the cheapest Claude model that works correctly. Three-tier progression: Haiku → Sonnet → Opus. Classify before responding. HAIKU (default): factual Q&A, greetings, reminders, status checks, lookups, simple file ops, heartbeats, casual chat, 1-2 sentence tasks. ESCALATE TO SONNET: code >10 lines, analysis, comparisons, planning, reports, multi-step reasoning, tables, long writing >3 paragraphs, summarization, research synthesis, most user conversations. ESCALATE TO OPUS: architecture decisions, complex debugging, multi-file refactoring, strategic planning, nuanced judgment, deep research, critical production decisions. Rule: If a human needs >30 seconds of focused thinking, escalate. If Sonnet struggles with complexity, go to Opus. Save 50-90% on API costs by starting cheap and escalating only when needed.
Why use this skill?
Optimize OpenClaw agent costs with smart model switching. Automatically route tasks between Haiku, Sonnet, and Opus to save 50-90% on your AI API bills.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/victorking2005/opusWhat This Skill Does
The smart-model-switching skill is an intelligent routing engine for the OpenClaw AI agent that dynamically assigns tasks to the most cost-effective and capable Claude model. By analyzing the complexity, depth, and technical requirements of a user prompt, the skill intelligently directs workloads across a three-tier hierarchy: Claude Haiku (for speed and low cost), Claude Sonnet (for balanced intelligence), and Claude Opus (for high-level reasoning). This ensures that simple tasks do not consume expensive resources while complex problems receive the necessary compute power, resulting in a 50-90% reduction in API operational costs.
Installation
To integrate this routing capability into your OpenClaw agent, use the following install command in your terminal:
clawhub install openclaw/skills/skills/victorking2005/opus
Ensure that you have your OpenClaw environment initialized and that you possess the necessary API credentials to access the Anthropic Claude model family.
Use Cases
- Efficient Triage: Automatically route status checks and greetings to Haiku, keeping your operations budget low.
- Automated Development: Scale up to Sonnet for writing features, debugging code, and generating documentation, and switch to Opus for architectural refactoring.
- Strategic Analysis: Utilize Opus for multi-file debugging or complex strategic planning that requires deep, nuanced judgment.
- Content Creation: Seamlessly transition from brief one-liner responses to long-form reports or articles without manual model swapping.
Example Prompts
- "Check the current status of the database heartbeat and summarize the logs." (Routes to Haiku)
- "Write a Python function to parse this JSON file and generate a table comparing the performance metrics." (Routes to Sonnet)
- "Analyze our current system architecture and propose a strategy for migrating to a microservices model with focus on scalability." (Routes to Opus)
Tips & Limitations
- The Golden Rule: Always evaluate if a human would take more than 30 seconds to think about the task; if yes, escalate from Haiku to Sonnet.
- Cost Awareness: Remember that Opus is 60x more expensive than Haiku. Use it strictly for tasks requiring high-level reasoning.
- Failure Recovery: If Sonnet returns an unsatisfactory result due to the complexity of a task, consider manually or programmatically escalating the specific sub-task to Opus.
- Scope: This skill is designed for model routing; ensure that your underlying agent configuration has the appropriate permissions for all three model tiers.
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-victorking2005-opus": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: external-api