Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计提醒

claude-oauth-renewalClaude OAuth renewal 搜索

Agent Skill

用于辅助安全审计、权限检查、凭据风险、认证流程和常见漏洞排查。它适合让 Agent 梳理敏感配置、检查依赖风险、分析鉴权逻辑或生成安全复核清单。使用时不能把工具输出直接当最终结论,涉及密钥、令牌、用户数据或生产系统时,应先确认最小权限、脱敏方式和操作边界。

总安装

12,896

周安装

532

GitHub Stars

公开资料未说明

下载量

4,213
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:claude-oauth-renewal(Claude OAuth renewal 搜索)
来源仓库:https://github.com/chenhab03/claude-oauth-renewal
安装命令:
openclaw skills install claude-oauth-renewal
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install claude-oauth-renewal

简介

通过心跳自动检测并更新过期的 Claude Code OAuth 令牌。 3 层更新:刷新令牌 → Chrome 浏览器自动化 → 用户警报。

SKILL.md

name
claude-oauth-renewal
description
Automatically detect and renew expired Claude Code OAuth tokens via heartbeat. 3-tier renewal: refresh token → Chrome browser automation → user alert.
homepage
https://github.com/anthropics/claude-code
metadata
{ "openclaw": { "emoji": "🔑", "requires": { "bins": ["claude", "security", "python3"], "platform": "macos" } } }

Claude Code OAuth Auto-Renewal

Automatically detect and renew expired Claude Code OAuth tokens during OpenClaw heartbeat cycles. Prevents agent downtime caused by token expiration.

When to Use

USE this skill when:

  • Your OpenClaw agent uses Claude Code as the AI provider
  • You want uninterrupted agent operation without manual token renewal
  • You're running OpenClaw on macOS with Chrome browser

How It Works

3-Tier Renewal Strategy

Heartbeat triggers check-claude-oauth.sh
  │
  ├─ Token healthy (>6h remaining) → silent exit ✓
  │
  ├─ Tier 1: claude auth status (refresh token)
  │   ├─ Success → silent exit ✓
  │   └─ Fail ↓
  │
  ├─ Tier 2: Browser automation (osascript + Chrome JXA)
  │   ├─ Start claude auth login
  │   ├─ Auto-click "Authorize" on claude.ai
  │   ├─ Extract auth code from callback page
  │   ├─ Feed code back to CLI via expect
  │   ├─ Success → silent exit ✓
  │   └─ Fail ↓
  │
  └─ Tier 3: Alert user → agent notifies via configured channel

Token Storage

Claude Code stores OAuth tokens in macOS Keychain under the service name Claude Code-credentials. The token JSON includes:

  • accessToken — API access token (prefix sk-ant-oat01-)
  • refreshToken — Used for automatic renewal (prefix sk-ant-ort01-)
  • expiresAt — Unix timestamp in milliseconds

Prerequisites

  1. macOS with security CLI (Keychain access)
  2. Claude Code installed and previously authenticated
  3. Google Chrome with View → Developer → Allow JavaScript from Apple Events enabled (for Tier 2)
  4. python3 available in PATH
  5. expect available (ships with macOS)

Setup

1. Copy the script

cp skills/claude-oauth-renewal/scripts/check-claude-oauth.sh scripts/check-claude-oauth.sh
chmod +x scripts/check-claude-oauth.sh

2. Add to HEARTBEAT.md

Add as the first step in your heartbeat execution:

## Execution Order

0. Run `bash scripts/check-claude-oauth.sh` — if output exists, relay as highest priority alert
1. (your other heartbeat checks...)

3. Test

# Normal check (silent if token healthy)
bash scripts/check-claude-oauth.sh

# Force trigger by setting high threshold
WARN_HOURS=24 bash scripts/check-claude-oauth.sh

Configuration

Environment VariableDefaultDescription
WARN_HOURS6Hours before expiry to start renewal attempts

Troubleshooting

"无法读取 Claude Code token"

  • Run claude auth login manually to establish initial credentials
  • Verify keychain access: security find-generic-password -s "Claude Code-credentials" -a "$(whoami)" -g

Tier 2 (browser automation) not working

  • Enable Chrome JXA: View → Developer → Allow JavaScript from Apple Events
  • Or via CLI: defaults write com.google.Chrome AppleScriptEnabled -bool true (restart Chrome)
  • Ensure you're logged into claude.ai in Chrome

JSON parsing errors

  • The script uses regex extraction (not json.loads) to handle truncated keychain output
  • If security -w truncates long values, the -g flag is used as fallback

Notes

  • Tier 1 (refresh token) handles most cases silently
  • Tier 2 (browser) is only needed when refresh token itself expires (typically weeks)
  • Tier 3 (alert) is the last resort when no automated renewal is possible
  • The script never stores or logs actual token values

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

73.69%
按下载量换算3,105

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

未展示

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills