skills.rest
freire19
freire19COMMUNITY
Explore all skills by this author →
GitHub

model-usage

Summarize CodexBar cost JSON into per-model usage and cost reports.

Updated May 8, 2026
Search Tags:#usage analytics#agent observability#codexbar#model breakdown#llm costs#json reporting#provider reporting
One-click install
npx skills add https://github.com/freire19/Mythos --skill model-usage-freire19
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: model-usage
Source: https://github.com/freire19/Mythos/tree/main/skills/model-usage
Command: npx skills add https://github.com/freire19/Mythos --skill model-usage-freire19

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

You can’t easily understand how much each LLM model costs your agent in practice, so you need a reliable way to summarize CodexBar local cost logs by provider and model.

Core Features & Use Cases

  • Per-model cost summaries: Produces “current model” (most recent daily entry) or “all models” breakdowns from CodexBar cost JSON.
  • Scriptable JSON output: Generates human-readable text or machine-friendly JSON for reports and automation.
  • Model selection logic: Chooses the highest-cost model from the most recent row’s modelBreakdowns, with fallbacks to modelsUsed when breakdowns are missing.
  • Use case: Compare which model is driving spend for a week’s agent runs, then standardize prompts or routing to reduce cost.

Quick Start

Ask for model-level usage/cost and run the bundled summary script to produce a current or full per-model report from CodexBar cost JSON.

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 LLM costs from CodexBar logs?▼

Per-model LLM costs are summarized by reading CodexBar cost JSON payloads and aggregating usage data by provider and model. The Skill parses modelBreakdowns to produce text or JSON reports for analysis.

What is the difference between current model and all models cost breakdowns?▼

Current model breakdowns estimate costs from the most recent daily entry in the cost JSON, selecting the highest-cost model. All models breakdowns aggregate complete usage and cost data across every stored cost JSON entry.

How do I generate JSON output for agent tooling cost reports?▼

JSON output for agent tooling cost reports is generated by applying optional formatting controls to the CodexBar cost JSON summary. The Skill produces machine-friendly JSON reports for automation and scriptable integration.

Can I filter provider reporting by specific models in CodexBar data?▼

Provider reporting can be filtered by specific models using optional provider and model controls. The Skill processes CodexBar cost JSON payloads to narrow summaries to targeted providers or models.

What happens when modelBreakdowns are missing from the cost JSON?▼

When modelBreakdowns are missing from the cost JSON, the Skill falls back to the modelsUsed field to determine model selection. This ensures cost summaries remain accurate even with incomplete breakdown data.

Related Skills

Explore More Data & Analytics Skills→
kcns008kcns008
5

model-usage

Summarize CodexBar cost logs by model for Codex and Claude.

Community
Intermediate
devCharuzudevCharuzu

llm-cost-optimizer

Optimize LLM API spend through model routing, caching, and cost observability.

Community
Advanced
felvieirafelvieira
22

cost-tracker

Aggregate tokens, API usage, and execution time to generate per-skill and session cost reports.

Community
Advanced
aresbitaresbit
46

model-usage

Analyze CodexBar cost logs to report per-model usage and costs.

Community
Advanced