token-budget-gate

What problem does it solve?

Multi-step and multi-agent AI tasks often silently consume far more tokens than expected, leading to unexpected cost overruns, session limit breaches, and wasted effort with no upfront warning.

Core Features & Use Cases

• Pre-execution cost estimation: Calculates expected token usage for planned tasks using file scope, agent count, step depth, and iteration factors.
• Tiered gate verdicts: Outputs Green/Yellow/Orange/Red verdicts with clear actions, from proceeding without comment to mandatory scope reduction before approval.
• Post-task calibration: Tracks actual vs estimated token usage to improve the accuracy of future cost estimates for the same task type.
• Use Case: A team running parallel agent dispatches across a large codebase can use this skill to flag high-cost runs upfront and automatically suggest narrowing file scope to stay within budget.

Quick Start

Ask the AI to run the token-budget-gate skill to estimate the token cost of your planned multi-agent code review task before starting work.