What This Skill Does
- Automatic Model Routing: Analyzes each task's complexity and routes it to the cheapest model that can handle it well.
- Token Budget Awareness: Tracks your token spend in real time and alerts you before you hit cost thresholds.
- Quality Guard Rails: Ensures responses still meet a quality floor, so cheaper never means worse output.
- Provider-Agnostic: Works across OpenAI, Anthropic, Google, and local models, always picking the best price-to-quality ratio.
- Spend Reports: Generates daily and weekly summaries of what you spent vs. what you would have spent without routing.
How to Install
Download the .md skill file from the link in your email, then copy it into your skills directory:
cp cost-optimizer.md ~/openclaw/skills/
Restart OpenClaw or run /skills reload and the Cost Optimizer will activate on your next task.