How do I install the Check skill?

Run npx skillmds@latest add emilfihlman/check in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Check safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Check?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Check free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

emilfihlman (@emilfihlman) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Check by emilfihlman — Agent Skill

Name: Check
Author: emilfihlman

Check

Query and summarize current Codex ChatGPT-plan usage windows, including used and remaining percentages, reset times, limit status, plan type, and usage-credit balance. Use when the user asks how much 5-hour or weekly Codex usage remains, whether a usage limit is near, or when a usage window resets. Use $usage:credits instead for banked reset-credit grants and expiry details.

emilfihlman Updated Sep 22, 2026

File contents

Usage Check

Inspect current Codex rate-limit windows without redeeming reset credits.

Workflow

Run scripts/show-codex-usage.sh.
If the sandbox blocks network access, rerun the same command with approval.
Report the primary 5-hour and secondary weekly windows, including used and remaining percentages, reset time, limit status, plan type, and usage-credit balance.
Route questions about banked reset-credit grants or expiry to $usage:credits.
Never print access tokens, account IDs, user IDs, or email addresses.

Script modes

Default: print a human-readable redacted summary.
--json: print redacted structured usage fields.
--raw: print the exact endpoint response. Use only when necessary and do not repeat identifiers or other sensitive fields in the response.

The helper accepts CODEX_AUTH_FILE and CODEX_USAGE_URL. It sends Codex credentials only to the default ChatGPT HTTPS host unless CODEX_ALLOW_UNSAFE_ENDPOINT=1 is explicitly set for controlled testing.

Treat failures as possible internal-endpoint, authentication, account, or rollout changes rather than user error.

Frequently asked questions