agent-expenditure

This Skill addresses the challenge of quantifying token waste in multi-agent parallel scheduling. In real-world development, when multiple AI agents are dispatched simultaneously to perform tasks, coordination overhead and redundant work can cause token consumption to spike, yet developers often rely on intuition to judge whether usage is excessive. This tool tracks each agent's token usage, calculates the coordination overhead ratio based on Brooks' Law, and automatically flags waste signals.

Usage is straightforward: after parallel agent scheduling is complete, simply trigger the analysis via a natural language command. It automatically reads the agent run logs, tallies each agent's token consumption and shared file conflict counts, and generates a coordination overhead report. You can then adjust subsequent scheduling strategies based on the report's recommendations (e.g., reducing the number of agents, optimizing task decomposition).

It is particularly well-suited for teams or individuals who frequently engage in multi-agent collaboration, especially those already experiencing token budget overruns or issues with duplicate agent outputs. For developers and operations engineers managing multiple parallel agent tasks, this tool provides data-driven decision-making support, helping to avoid blindly increasing the agent count.

It is recommended to use this tool for retrospective analysis after each large-scale parallel scheduling session. Note that it is primarily designed for post-hoc analysis and cannot intervene in real-time during agent execution. Additionally, ensure that agent logs include token consumption and file operation records. For single-agent workflows, this tool will not produce meaningful data.