Token Budget Guard
Stop burning context. Manage your agent's token budget intelligently.
The Problem
AI agents waste 40-60% of tokens on:
- Repeatedly loading full schemas when summaries suffice
- Including irrelevant context from previous turns
- Not compressing before context window fills
- Loading entire files when snippets would do
The AAI Gateway showed 99% token savings are possible. This skill makes token budgeting automatic.
When to Use
- "token budget", "reduce tokens",
[Description truncada. Veja o README completo no GitHub.]