model-usage

Summarize CodexBar local cost logs by model with day-range filtering.

1|Updated May 12, 2026
One-click install
npx skills add https://github.com/estebanrfp/gos --skill model-usage-estebanrfp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/estebanrfp/gos/tree/main/skills/model-usage
Command: npx skills add https://github.com/estebanrfp/gos --skill model-usage-estebanrfp

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill removes the need to manually inspect CodexBar cost logs when you want to understand which model was used and how much it cost.

Core Features & Use Cases

  • Current Model Summary: Identifies the most recent active model from local cost data and reports its latest and total spend.
  • Full Model Breakdown: Aggregates costs across all models for Codex or Claude and can limit results to a recent day range.
  • Use Case: A team lead reviewing monthly AI spend can quickly see which model drove costs, compare providers, and spot expensive usage patterns without parsing raw JSON.

Quick Start

Ask for a summary of the current CodexBar model usage from your local cost data.

Frequently Asked Questions about model-usage

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

FAQPage Schema
How do I aggregate local CodexBar cost logs by AI model?

You aggregate local CodexBar cost logs by AI model by parsing local JSON cost output to generate a full model breakdown, applying optional day-range filtering to isolate recent Codex and Claude usage spend.

What is the best way to see per-model AI costs at a glance from local logs?

Seeing per-model AI costs at a glance from local logs involves summarizing CodexBar JSON output to report the latest and total spend for the most recent active model without parsing raw data manually.

Can I limit a Claude usage report to a specific day range?

You can limit a Claude usage report to a specific day range by applying day-range filtering to your local CodexBar cost logs, restricting the model breakdown aggregation to only include recent usage data.

Do I need local JSON files to analyze Codex and Claude usage costs?

You need local CodexBar JSON cost output to analyze Codex and Claude usage costs, as the summarization process relies on parsing these local files to aggregate model breakdowns and current-model snapshots.

What happens if model breakdowns are missing from the CodexBar cost output?

When model breakdowns are missing from the CodexBar cost output, the analysis applies fallback logic to ensure it can still summarize the current model snapshot and total spend from the available local logs.