model-usage

Summarize per-model costs from CodexBar JSON cost logs.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/MaxSaiets/myOpenCL --skill model-usage-maxsaiets
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/MaxSaiets/myOpenCL/tree/main/openclaw/skills/model-usage
Command: npx skills add https://github.com/MaxSaiets/myOpenCL --skill model-usage-maxsaiets

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

CodexBar local cost logs can be difficult to parse and summarize into per-model usage costs; this Skill provides a focused way to extract per-model costs for Codex or Claude, including the current model and a full breakdown.

Core Features & Use Cases

  • Current model view: quickly identify the most expensive model in the latest data.
  • All-models breakdown: compute total cost per model from daily breakdowns.
  • Flexible inputs & outputs: accepts a local cost JSON file or stdin and returns either human-readable text or JSON suitable for automation.

Quick Start

Run the bundled Python script to summarize per-model costs from a local 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 local logs?

You can summarize per-model cost usage by parsing CodexBar local JSON cost logs via a Python script to extract per-model totals. It identifies the current highest-cost model and processes daily breakdowns for Codex or Claude.

What is the best way to identify the current highest-cost model in my CodexBar logs?

Identifying the highest-cost model involves parsing local CodexBar JSON cost logs to extract current model expenses. The script automatically isolates the most expensive model from the latest payload data for quick review.

Can I use stdin to pass JSON cost payloads for model usage analysis?

Yes, you can pass JSON cost payloads via stdin for model usage analysis. The tool accepts input from a local file or stdin and returns either human-readable text or structured JSON results suitable for automation.

Does the model usage summarizer handle daily breakdowns for both Codex and Claude?

Yes, the summarizer handles daily breakdowns for both Codex and Claude. It parses daily JSON cost payloads to compute per-model totals, applying the analysis to all-model breakdowns and current model views.

Are there limitations when parsing local JSON cost logs for per-model totals?

The primary limitation is that parsing local JSON cost logs requires correctly structured CodexBar payload data. The tool processes local files or stdin exclusively and does not fetch logs from remote APIs or handle non-JSON formats.