home / skills / openclaw / skills / model-usage

model-usage skill

safe

This skill summarizes per-model usage and costs from CodexBar for Codex or Claude, returning current or full model breakdowns to inform budgeting.

npx playbooks add skill openclaw/skills --skill model-usage

Review the files below or copy the command above to add this skill to your agents.

Files (4)

SKILL.md

2.0 KB

---
name: model-usage
description: Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.
metadata: {"clawdbot":{"emoji":"📊","os":["darwin"],"requires":{"bins":["codexbar"]},"install":[{"id":"brew-cask","kind":"brew","cask":"steipete/tap/codexbar","bins":["codexbar"],"label":"Install CodexBar (brew cask)"}]}}
---

# Model usage

## Overview
Get per-model usage cost from CodexBar's local cost logs. Supports "current model" (most recent daily entry) or "all models" summaries for Codex or Claude.

TODO: add Linux CLI support guidance once CodexBar CLI install path is documented for Linux.

## Quick start
1) Fetch cost JSON via CodexBar CLI or pass a JSON file.
2) Use the bundled script to summarize by model.

```bash
python {baseDir}/scripts/model_usage.py --provider codex --mode current
python {baseDir}/scripts/model_usage.py --provider codex --mode all
python {baseDir}/scripts/model_usage.py --provider claude --mode all --format json --pretty
```

## Current model logic
- Uses the most recent daily row with `modelBreakdowns`.
- Picks the model with the highest cost in that row.
- Falls back to the last entry in `modelsUsed` when breakdowns are missing.
- Override with `--model <name>` when you need a specific model.

## Inputs
- Default: runs `codexbar cost --format json --provider <codex|claude>`.
- File or stdin:

```bash
codexbar cost --provider codex --format json > /tmp/cost.json
python {baseDir}/scripts/model_usage.py --input /tmp/cost.json --mode all
cat /tmp/cost.json | python {baseDir}/scripts/model_usage.py --input - --mode current
```

## Output
- Text (default) or JSON (`--format json --pretty`).
- Values are cost-only per model; tokens are not split by model in CodexBar output.

## References
- Read `references/codexbar-cli.md` for CLI flags and cost JSON fields.

Overview

This skill summarizes per-model usage and cost from CodexBar local cost logs for Codex or Claude. It supports a "current model" view (most recent daily entry) and a full breakdown across all models. Use it when you need a scriptable, human- or machine-readable summary of model-level costs from CodexBar JSON output.

How this skill works

The script reads CodexBar cost JSON produced by the CLI or from a file/stdin. For the current-model mode it selects the most recent daily row with modelBreakdowns and picks the highest-cost model, falling back to the last modelsUsed entry if breakdowns are missing. In all-models mode it aggregates cost values per model and emits text or JSON output.

When to use it

You need a quick per-model cost snapshot for the most recent billing day.
You want a full breakdown of costs across all models in your CodexBar data.
You need output that can be consumed by other scripts (JSON or plain text).
You want to override automatic detection and force a specific model summary.
You have CodexBar cost JSON saved to a file or streaming via stdin.

Best practices

Fetch fresh cost JSON using the CodexBar CLI before running the script to ensure up-to-date results.
Use --mode current for quick checks and --mode all for audits or reports.
Pass --input - to stream JSON via stdin for automated pipelines.
Use --format json --pretty when integrating with downstream tools that expect structured data.
Explicitly set --model <name> when you need a deterministic model selection instead of the top-cost heuristic.

Example use cases

Run a daily job that pipes codexbar cost --format json into the script to extract the current most expensive model.
Generate a monthly report that aggregates costs per model by running the script in --mode all and saving pretty JSON.
Automate alerts that notify when a specific model's daily cost exceeds a threshold by combining the script with simple filtering.
Quickly inspect model-level spend after a deployment to verify new model usage is within budget.
Include the script in CI/CD pipelines to produce a compact per-model cost summary as part of billing checks.

FAQ

What input sources are supported?

The script supports running the CodexBar CLI directly, reading a JSON file via --input, or accepting JSON on stdin.

How does current-mode pick the model?

It uses the most recent daily row with modelBreakdowns and selects the highest-cost model, falling back to the last modelsUsed entry if breakdowns are missing; you can override with --model.