agent-expenditure

This Skill addresses the issue of surging and hard-to-quantify token consumption due to coordination overhead in multi-agent parallel scheduling scenarios. In practice, developers often need to invoke multiple AI agents concurrently to handle complex tasks but lack effective tools to track each agent's token usage. This makes it difficult to assess whether adding more agents delivers proportional value, often leading to resource waste.

Usage is straightforward: simply trigger this Skill for analysis after completing a parallel agent scheduling task. It automatically calculates each agent's token consumption, identifies coordination overhead caused by factors like file read/write conflicts and duplicated work, and generates a detailed report incorporating guidance based on Brooks's Law. You can then use the report's "waste signals" and "post-scheduling checklist" to optimize future scheduling strategies.

It is well-suited for teams or individuals who need to coordinate multi-agent tasks, engage in complex project development, or perform data analysis—particularly those already using parallel agent patterns but struggling with uncontrolled costs and unclear efficiency gains. Product managers and project managers can also leverage it to evaluate the return on investment of multi-agent workflows, enabling more informed resource allocation decisions.

We recommend using this tool for review after each parallel scheduling task involving four or more agents. Note that it primarily performs post-hoc analysis and is not suitable for single-agent workflows or real-time intervention during agent execution. Additionally, it focuses on coordination efficiency analysis, not overall token budget planning, which should be handled by dedicated token budget management tools.