model-usage

Aggregate per-model cost usage from CodexBar cost JSON.

Updated Mar 13, 2026
One-click install
npx skills add https://github.com/Yogi-076/Vajrascan.on --skill model-usage-yogi-076
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/Yogi-076/Vajrascan.on/tree/main/moltbot/skills/model-usage
Command: npx skills add https://github.com/Yogi-076/Vajrascan.on --skill model-usage-yogi-076

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Provides a quick, scriptable way to extract and summarize per-model cost usage from CodexBar local cost logs so users can identify which models drive spending and view recent model-specific costs.

Core Features & Use Cases

  • Per-model cost aggregation: Aggregates modelBreakdowns across daily rows to compute total cost per model.
  • Current vs. all models: Detects the current (most recent/highest-cost) model or produces a full model breakdown across the dataset.
  • Flexible input sources: Accepts codexbar CLI output, a JSON file path, or JSON via stdin and outputs human-readable text or structured JSON for scripting.
  • Use case: Useful for developers and cost-ops engineers checking which Codex or Claude models are incurring the most local cost during audits or monthly reporting.

Quick Start

Run the model-usage script to summarize Codex or Claude cost data by invoking codexbar cost and feeding the JSON output to the script or by passing a saved codexbar cost JSON file.

Frequently Asked Questions about model-usage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I summarize per-model cost usage from CodexBar logs?

Summarizing per-model cost usage involves aggregating modelBreakdowns across daily rows to compute total cost per model. The script accepts codexbar CLI output, a JSON file, or stdin.

How do I check which Claude or Codex model is driving the most local cost?

To check which Claude or Codex model drives the most local cost, detect the current model by identifying the most recent or highest-cost entry. The script processes local cost logs to pinpoint top spending models.

Can I pass codexbar cost JSON directly through stdin for model breakdown?

Yes, you can pass codexbar cost JSON directly through stdin for model breakdown. The script accepts flexible input sources including codexbar CLI output, a JSON file path, or JSON via stdin to aggregate usage data.

Does the model-usage script output structured JSON for automated cost reporting?

Yes, the model-usage script outputs structured JSON for automated cost reporting. It supports both human-readable text and structured JSON output formats, making it suitable for scripting and monthly cost audits.

What is the difference between current mode and all mode when aggregating model costs?

The difference between current mode and all mode is that current mode detects the single most recent or highest-cost model, while all mode produces a full model breakdown across the entire dataset.

Do I need any external dependencies to run the CodexBar cost aggregation script?

No external dependencies are required to run the CodexBar cost aggregation script. It operates independently with no listed dependencies, processing local Codex and Claude cost logs.