xAI OAuth Usage Checker
Read-only Hermes Agent skill for checking xAI OAuth usage, remaining quota, and reset time with Japanese/English output.
hirory
@hirorylabo
Install
$ openclaw skills install @hirorylabo/xai-oauth-usageHermes xAI Usage Checker
Use Hermes' canonical OAuth credential to check xAI weekly usage and reset time in read-only mode. Do not modify credentials, reauthenticate, perform billing actions, or claim that the reported quota is model-specific.
When to Use
- When a user asks for xAI OAuth or Grok usage, remaining quota, or reset time
- When
xai-oauth/grok-4.6reportsbillingor quota exhaustion - When deciding whether Hermes fallback behavior or
x_searchcan be retested after the quota resets
Prerequisites
- A valid
xai-oauthcredential in Hermes' canonical auth store python3and HTTPS access tohttps://cli-chat-proxy.grok.com- User approval for a read-only OAuth request to the xAI Billing endpoint; ask first if approval has not been given
How to Run
Run the script through Hermes' terminal tool with sanitized JSON output:
terminal(command='python3 "${HERMES_SKILL_DIR}/scripts/check_xai_usage.py" --json', timeout=30)
For human-readable output, use the same terminal call without --json.
Human-readable output and stderr follow HERMES_LANGUAGE, then Hermes' display.language. Japanese settings produce Japanese output; every other value and configuration lookup failure produce English output. JSON keys and JSON error strings always remain in English.
Entry point: scripts/check_xai_usage.py
Quick Reference
| Field | Meaning |
|---|---|
used_percent | Current weekly usage percentage |
remaining_percent | 100 - used_percent, clamped at zero |
period_end_jst | Billing API period end converted to JST |
remaining_seconds | Seconds from execution time until the period ends |
product_usage | Per-product usage returned by the Billing API |
source_host | cli-chat-proxy.grok.com on success |
Procedure
- Confirm that the request is a read-only quota check and does not include credential changes or billing actions.
- Run the
terminalcall in How to Run once. Do not redirect, log in, log out, refresh, reset, or top up. - If
status: ok, report only usage, remaining quota,period_end_jst, approximate time remaining, andproduct_usage, using Hermes' display language. - If
remaining_percentis zero andremaining_secondsis positive, treat the quota as not yet recovered. - If
status: error, report onlyerrorandhttp_status. Do not automatically reauthenticate after a 401 or 403 response. - Never display, save, or include access tokens, refresh tokens, ID tokens,
auth.jsoncontents, or the raw API response in an answer.
Example report:
xAI OAuth weekly quota is 100% used.
Reset: 2026-08-29 23:04 JST, approximately 29 hours 3 minutes remaining.
Breakdown: GrokBuild 100%.
Pitfalls
- The
billingperiod may represent a shared weekly quota. Do not claim it is aGrok 4.6-specific limit. - Remaining time decreases while the command runs. Treat
period_end_jstas authoritative and label remaining time as approximate. - If Hermes still reports
billingafter the reset time, query once more instead of performing a blind reset. - Prefer the API result even when no usage page is visible, but report the value as unverified if the request fails.
- The script rejects redirects. Do not replace this behavior with an implementation that sends the Authorization header to another host.
Verification
In a source checkout, run the network-free regression tests from the skill directory:
python3 -m unittest discover -s tests -v
For a skill installed through Hermes, run this network-free smoke test:
terminal(command='python3 "${HERMES_SKILL_DIR}/scripts/check_xai_usage.py" --help', timeout=30)
Only with user approval, run the live request from How to Run and verify status: ok, the expected source_host, and the absence of token fields.
日本語での概要
HermesのxAI OAuth週間利用率・残量・リセット時刻を、認証情報を変更せず読み取り専用で確認するskillです。 通常表示とstderrはHermesが日本語設定なら日本語、それ以外は英語になります。JSONのキーとエラー文は英語固定です。 実行前にユーザーの承認を確認し、tokenや生のAPIレスポンスは表示・保存しません。
Top skills in this category
Skill Vetter
@spclaudehomeSecurity-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
Multi Search Engine
@gpyangyoujunMulti search engine integration with 16 engines (7 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and Wolfra...
API Gateway
@byungkyuCall third-party APIs through the Maton gateway, which injects the credential for an app the user has already connected. Use this skill when the user names a connected app and a concrete action in it - read a mailbox, query a CRM, file an issue, update a spreadsheet, run a query through a connected
MoltGuard - Security & Antivirus & Guardrails
@thomaslwangMoltGuard — OpenClaw security guard by OpenGuardrails. Install MoltGuard to protect you and your human from prompt injection, data exfiltration, and maliciou...
1password
@steipeteSet up and use 1Password CLI (op). Use when installing the CLI, enabling desktop app integration, signing in (single or multi-account), or reading/injecting/running secrets via op.