Back to skills
token-economy
by pfaria32
Coding Agents & IDEs—Not yet rated
Your rating
Brief
Reduce OpenClaw costs by 60-80% through intelligent model routing and context management.
Description
What This Skill Does
Reduces agent costs by 60-80% through model routing and context management. Routes tasks to the cheapest capable model first (GPT-4o), escalating to Sonnet or Opus only when needed. Eliminates heartbeat API calls when there's nothing to process, and enforces a daily budget cap.
When to Use It
- Routing simple tasks to cheaper models automatically
- Eliminating wasted API calls from empty heartbeat checks
- Enforcing a daily spending cap to prevent runaway costs
- Auditing token usage with per-request logging
- Auto-switching back to a cheaper model during idle periods
Installation
clawhub install pfaria32/token-economyRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.