Nm Conserve Agent Expenditure

>- Track per-agent token usage and flag waste patterns in parallel dispatch workflows. Consult after running parallel agents to evaluate whether expenditure...

Security Scan

The skill's requirements and instructions are coherent with its stated purpose (post‑dispatch token-waste analysis); it is instruction-only, requests no credentials, and does not attempt unexplained system or network access.

Purpose & Capability

Name/description (monitor per-agent token usage and flag waste patterns) align with the provided guidance and the included 'waste-signals' module. The skill does not request unrelated binaries, credentials, or config paths beyond the waste-signals doc. Minor metadata inconsistencies noted: SKILL.md reports version 1.8.2 while registry metadata lists 1.0.0; the declared required config path 'night-market.modules/waste-signals.md' does not exactly match the file manifest path 'modules/waste-signals.md' — likely a namespacing mismatch but not a functional security issue.

Instruction Scope

SKILL.md content is narrowly scoped to post-dispatch analysis: waste signal detection criteria, checklists, and mitigation suggestions. It references other Night Market rules and skills (plan-before-large-dispatch, token-conservation, conjure:agent-teams) but does not instruct reading arbitrary system files, environment variables, or exfiltrating data. All file references are to project/skill artifacts rather than user secrets.

Install Mechanism

Instruction-only skill with no install spec and no code files — lowest-risk install profile. Nothing is downloaded or written to disk by an installer in the manifest.

Credentials

No required environment variables, no primary credential, and no requests for unrelated secrets. The only required config is the waste-signals module, which is included. There is no disproportionate credential access.

Persistence & Privilege

always is false, the skill is user-invocable and allows model invocation (platform default). It does not request persistent system-level presence, nor does it modify other skills' configs. Autonomous invocation is normal here and not combined with other red flags.

Assessment

This skill appears to do what it says: post-hoc analysis of agent token usage with five clear waste-signal definitions. Before installing, confirm that the declared config path (night-market.modules/waste-signals.md) maps to the module file you have (manifest shows modules/waste-signals.md) and that the referenced cross-skill rules are available in your environment. Because it can be invoked by the agent, avoid combining it with other skills that request broad credentials; on its own it does not ask for secrets or perform I/O outside its documented scope.

Night Market Skill — ported from claude-night-market/conserve. For the full experience with agents, hooks, and commands, install the Claude Code plugin.

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 Count	Expected Overhead	Guidance
1-3	Negligible	Dispatch freely
4-5	10-15%	Acceptable; plan first
6-8	20-30%	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

• .claude/rules/plan-before-large-dispatch.md for the 4+ agent planning requirement
• conserve:token-conservation for session-level token budgeting
• conjure:agent-teams for dispatch coordination