mcpField
A
agent-expenditure

agent-expenditure

github.com/atholaAgent workflows34384 installsUpdated 1d ago
Repository

About agent-expenditure

A post-dispatch review skill for assessing token waste and coordination overhead in parallel agent runs. It provides review questions and waste signals to guide future agent counts; it is intended for completed multi-agent work, not active runs or single-agent workflows.

Key features

  1. 01Review agent findings for duplication
  2. 02Assess token expenditure against value
  3. 03Check coordination overhead and waste signals
  4. 04Recommend future agent counts

Use cases

  1. 01Review a completed parallel agent run
  2. 02Decide whether to increase agent count
  3. 03Retrospect on agent-heavy workflows

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/athola/claude-night-market --skill agent-expenditure

After installing, type/agent-expenditurein Claude Code to confirm it loaded.

Bundled filesmodules/waste-signals.md

Details

From the project README

Agent Token Waste Monitoring

When To Use

  • After parallel agent dispatch completes
  • When evaluating whether to increase agent count
  • During retrospectives on agent-heavy workflows
  • When plan-before-large-dispatch rule triggers

When NOT To Use

  • Single-agent workflows (no coordination overhead)
  • During active agent execution (post-hoc analysis)
  • For token budgeting (use token-conservation instead)

Brooks's Law for Agents

Dispatching more agents does not always help. Coordination overhead grows with agent count:

Agent CountExpected OverheadGuidance
1-3NegligibleDispatch freely
4-510-15%Acceptable; plan first
6-820-30%Monitor closely
9+30%+Likely counterproductive

Coordination overhead is measured as shared-file conflicts: concurrent Read/Write operations on the same file by different agents, as a percentage of total agent runtime.

Post-Dispatch Review Checklist

After parallel agent runs, evaluate:

  1. Did each agent produce unique findings?
  2. Was total token expenditure proportional to value?
  3. Did any agent duplicate another's work?
  4. Would fewer agents have produced the same result?

If 2+ questions answer no, reduce agent count in future dispatches of the same type.

Waste Signals

See modules/waste-signals.md for the 5 waste signal categories and detection criteria.

Cross-References

  • Dispatching 4 or more agents needs an agreed plan first: the roster, each agent's scope, and the output contract it returns. 1-3 agents may dispatch directly.
  • conserve:token-conservation for session-level token budgeting
  • conjure:agent-teams for dispatch coordination

Exit Criteria

  • All 4 post-dispatch review questions answered with explicit yes/no per agent (unique findings, proportional expenditure, no duplication, fewer agents sufficient)
  • Waste signals from modules/waste-signals.md checked against the completed run; any triggered signal named with the category
  • A recommendation is produced: either "reduce agent count to N" or "dispatch was efficient" with coordination overhead percentage
  • If 2+ review questions answer no, a concrete agent-count reduction is stated for future dispatches of the same type

Comments

Markdown supported