model-usage

Extract per-model usage costs from CodexBar cost logs into JSON summaries.

Updated May 7, 2026

Search Tags:#codex #claude #cost-tracking #model-usage #codexbar #usage-summary #ai-spend

Ready to run

npx skills add https://github.com/MOODMNKY-LLC/mnky-atlas --skill model-usage-moodmnky-llc

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/MOODMNKY-LLC/mnky-atlas/tree/main/skills/model-usage
Command: npx skills add https://github.com/MOODMNKY-LLC/mnky-atlas --skill model-usage-moodmnky-llc

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Manually parsing raw CodexBar cost logs to identify which AI models (Codex or Claude) are driving the highest usage costs is tedious and time-consuming, especially for teams tracking AI API spend.

Core Features & Use Cases

Per-Model Cost Summarization: Extracts and aggregates usage costs by individual model from CodexBar local cost logs for both Codex and Claude providers.
Flexible Breakdown Options: Supports both current most-recent model cost breakdowns and full historical all-models cost summaries.
Use Case: Development teams can quickly identify expensive AI model usage patterns without manually sifting through raw log data, helping optimize AI tooling spend.

Quick Start

Use the model-usage skill to generate a full per-model cost breakdown for all Claude usage from your local CodexBar logs.

Frequently Asked Questions about model-usage

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

FAQPage Schema

How do I track AI API spend for Codex and Claude models without parsing raw logs?▼

You can track AI API spend by using a bundled Python script to process CodexBar local cost logs, extracting and aggregating per-model usage costs into a human-readable cost summary for Codex and Claude providers.

What is the best way to get a per-model cost breakdown from CodexBar logs?▼

The best way to get a per-model cost breakdown is to process the raw cost JSON output from CodexBar logs, which extracts and aggregates usage costs by individual AI model for both current and historical summaries.

Does summarizing AI model usage costs require any specific environment setup?▼

Summarizing AI model usage costs requires the CodexBar CLI to be installed and accessible in your environment, as the Skill relies on this tool to retrieve the raw cost logs needed for processing.

Can I view historical all-models cost summaries as well as recent model breakdowns?▼

Yes, you can view both current most-recent model cost breakdowns and full historical all-models cost summaries, allowing you to analyze expensive AI model usage patterns over time for individual developers and teams.

Why does manually parsing CodexBar cost logs make tracking AI spend difficult?▼

Manually parsing CodexBar cost logs is tedious and time-consuming because raw cost JSON output lacks aggregation, making it difficult to identify which Codex or Claude AI models are driving the highest usage costs.

Do I need to manually sift through raw JSON data to optimize AI tooling spend?▼

No, you do not need to manually sift through raw JSON data to optimize AI tooling spend; the processing logic automatically converts raw cost logs into structured JSON or human-readable cost summaries.