model-usage

Summarize CodexBar cost logs by model with per-day aggregation and date filtering.

Updated Aug 13, 2025
One-click install
npx skills add https://github.com/JoeyJoziah/investment-analysis-platform --skill model-usage-joeyjoziah
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/JoeyJoziah/investment-analysis-platform/tree/main/.claude/skills/model-usage
Command: npx skills add https://github.com/JoeyJoziah/investment-analysis-platform --skill model-usage-joeyjoziah

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

It eliminates the manual work of figuring out which Codex or Claude model is driving local usage cost, and turns raw CodexBar cost logs into a clear per-model summary.

Core Features & Use Cases

  • Current Model Detection: Identifies the most recent active model from daily cost rows, with fallback logic when detailed breakdowns are missing.
  • Per-Model Cost Aggregation: Totals spend across all recorded days so you can compare model usage and budget impact.
  • Flexible Input Handling: Reads from CodexBar directly, from a saved JSON file, or from stdin for scripting and automation.
  • Use Case: Review yesterday’s Claude spend, compare all Codex models over the last 30 days, or generate a JSON report for downstream tooling.

Quick Start

Use the model-usage skill to summarize a CodexBar cost JSON file and show the current model or full per-model spend breakdown.

Frequently Asked Questions about model-usage

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

FAQPage Schema
How do I summarize model usage spend from CodexBar logs?

To summarize model usage spend from CodexBar logs, you parse daily cost rows to aggregate totals per model. This process identifies the current active model and compares historical budget impact using text or JSON outputs.

What is the best way to break down Claude and Codex costs by model?

The best way to break down Claude and Codex costs by model is to apply per-day aggregation to local cost JSON workflows. This compares model usage by totaling spend across all recorded days with optional date filtering.

Can I read CodexBar cost data from a saved JSON file or stdin?

Yes, you can read CodexBar cost data directly, from a saved JSON file, or from stdin. This flexible input handling supports parsing daily rows and model breakdowns for scripting and automation workflows.

How does current model detection work when detailed cost breakdowns are missing?

Current model detection identifies the most recent active model from daily cost rows and applies fallback model logic when detailed breakdowns are missing. This ensures accurate usage reporting even with incomplete local cost logs.

How do I generate a JSON report of historical Codex model usage?

To generate a JSON report of historical Codex model usage, aggregate spend across recorded days with optional date filtering. This produces a structured JSON summary for downstream tooling and budget analysis.