model-usage

Extract and summarize per-model usage costs from CodexBar local cost logs.

55|12|Updated Mar 14, 2026
One-click install
npx skills add https://github.com/Yapie0/safe-openclaw --skill model-usage-yapie0
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/Yapie0/safe-openclaw/tree/main/skills/model-usage
Command: npx skills add https://github.com/Yapie0/safe-openclaw --skill model-usage-yapie0

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

CodexBar logs scattered across local environments make it hard to understand which models drive costs. This Skill consolidates per-model usage costs from local logs for quick, actionable insights.

Core Features & Use Cases

  • Local aggregation of per-model usage costs from CodexBar logs (JSON) to reveal model-level spending.
  • Supports current-model view and all-model breakdowns for Codex or Claude providers.
  • Outputs human-readable text or structured JSON with optional date-range filtering.

Quick Start

Summarize the per-model costs from CodexBar using the most recent daily entry.

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 usage costs from CodexBar local logs?

To summarize per-model usage costs from CodexBar logs, you can use a tool that extracts and aggregates local JSON cost data. It processes these logs to provide a clear model-level spending view for quick insights.

Can I filter model-level cost breakdowns by a specific date range?

Yes, you can filter model-level cost breakdowns by a specific date range. By applying a days parameter, the summary restricts the extracted usage costs to only include entries from within that defined period.

Does this approach support cost breakdowns for both Codex and Claude providers?

Yes, this approach supports cost breakdowns for both Codex and Claude providers. You can generate a current-model view or a full-model breakdown, validating the specific provider input to aggregate the correct logs.

What is the best way to view per-model costs if CodexBar is not installed?

The best way to view per-model costs without CodexBar installed is to provide a JSON payload directly via input. The system validates this payload for provider, mode, and days to output the cost summary.

Can I get structured JSON output instead of text for my model usage costs?

Yes, you can get structured JSON output instead of text for your model usage costs. The summarization process supports both human-readable text and structured JSON formats to integrate with downstream analytics.