model-usage

Aggregate per-model costs from local CodexBar JSON logs by provider.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/qkal/superbyte --skill model-usage-qkal
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/qkal/superbyte/tree/main/skills/model-usage
Command: npx skills add https://github.com/qkal/superbyte --skill model-usage-qkal

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

CodexBar generates local cost logs by model, and this skill summarizes those logs to reveal per-model spending for Codex and Claude.

Core Features & Use Cases

  • Per-model cost summaries from local CodexBar logs for quick visibility into spending.
  • Supports both current-model and all-model views to fit budgeting and forecasting needs.
  • Works with local JSON log data, enabling day-range filtering and provider-specific insights.

Quick Start

Run the model_usage tool to summarize your local CodexBar usage by model for Codex or Claude.

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 costs from local CodexBar logs?

You can summarize per-model CodexBar usage locally by reading local JSON cost logs and aggregating model costs by provider for Codex and Claude. The skill exposes current and all-model summaries with optional day-range filtering.

Can I filter local log analysis by a specific day range for Codex and Claude?

Yes, local log analysis supports day-range filtering. The skill reads local CodexBar cost JSON logs and applies day-range filters to generate provider-specific historical cost summaries for Codex and Claude data.

What is the difference between current-model and all-model views in cost analysis?

Current-model views summarize spending for the active model, while all-model views aggregate costs across every model. Both are generated from local CodexBar JSON logs to fit budgeting and daily forecasting needs.

Does this approach require any external dependencies to analyze local JSON cost logs?

No external dependencies are required to analyze local JSON cost logs. The skill operates independently using local scripts and references to read and aggregate CodexBar cost data directly on your machine.

How do I track daily spending for Codex and Claude models on my local machine?

You track daily spending by running the model_usage tool against local CodexBar logs. It identifies per-model costs for Codex and Claude, enabling daily and historical cost analysis directly on local machines.