# Openai Usage

> Report current OpenAI usage/rate-limit health from Codex ChatGPT limits.

- Skill: `diegosouzapw/openai-usage` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add diegosouzapw/openai-usage`
- Raw SKILL.md: https://api.skillmd.com/api/skills/diegosouzapw/openai-usage/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Docs & Writing
- Author: diegosouzapw (https://skillmd.com/u/diegosouzapw)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/diegosouzapw/openai-usage

---


# OpenAI Usage Skill

Use this skill when the user asks things like:
- "what is my OpenAI usage at?"
- "show my OpenAI limits"
- "how much OpenAI budget do I have left?"

## Command

- Human-readable output:
  - `node skills/openai-usage/check-openai-usage.js`
- JSON output:
  - `node skills/openai-usage/check-openai-usage.js --json`

## Data Source

- Codex app-server `account/rateLimits/read` (ChatGPT mode)
- No manual fallback. If Codex rate limits are unavailable, return an actionable error.

## Behavior

- Reports 5-hour and weekly remaining percentages.
- Computes one health stage (`healthy`, `medium`, `low`, `critical`) using separate 5-hour and weekly thresholds.
- Stage rules:
  - `critical` if 5-hour <= 10% OR weekly <= 10%
  - `low` if 5-hour <= 25% OR weekly <= 20%
  - `medium` if 5-hour <= 50% OR weekly <= 30%
  - otherwise `healthy`