安装方式
手动下载安装
下载 ZIP 后解压到技能目录即可安装。若在桌面客户端 WebView中直接下载出现异常,本站会改为提示页 + 原始链接,请按页内说明操作。
下载 ZIP (shub-jwt-v1.0.1.zip)触发指令
/jwt
跨平台安装指引
该技能声明兼容以下 1 个平台,将 ZIP 解压到对应目录即可被识别。
unzip shub-jwt-v1.0.1.zip -d ~/.claude/skills/
mkdir -p 创建;启用 Skill 后请重启对应 Agent 让配置生效。
使用指南
JWT 实践(通用)
围绕 JWT 实践(通用):签发、校验、刷新与密钥轮换;算法与 clock skew 处理见 ZIP,与「jwt-toolkit」可对照。 无需在每次任务前把零散英文说明手工拼进上下文,也 减少 与客户端默认行为脱节的试错;具体命令、钩子与 JSON 参数仍以 ZIP 包内 SKILL.md 为权威。下文结构与站内 MCP CLI 类专题稿相同:何时用、前置、流程、速查与故障。
何时使用
- 签发、校验、刷新与密钥轮换
- 算法与 clock skew 处理见 ZIP,与「jwt-toolkit」可对照
- 已获取本技能 ZIP,并准备在 Claude Code / OpenClaw 中按 SKILL.md 挂载。
- 希望用中文专题稿快速判断「该不该启用」,再深入英文 SKILL 查参数与边界。
- 需要与团队对齐同一套触发方式、目录约定或回调格式时。
前置条件
- 通用:可运行 Claude Code 或文档要求的客户端;有可读写的项目工作区(或 SKILL.md 指定的沙箱目录)。
- 权威细节:API Key / OAuth、钩子路径、环境变量以 ZIP 内 SKILL.md 为准。
典型流程
- 从 ClawHub / 站内分发获取技能 ZIP,校验版本与校验和(若提供)。
- 阅读 SKILL.md 的安装段落:目录落点、客户端类型(Claude Code / OpenClaw / 脚本)。
- 用文档中的最小示例完成第一次调用(单文件修改、单次查询或单次委派)。
- 确认工作目录、权限边界与输出路径后,再处理多文件或长耗时任务。
- 需要回调 / Webhook / 通知时,按 SKILL.md 配置端点并在测试环境先验通。
与 ZIP / SKILL.md 的关系
站内专题稿与 MCP CLI 类 oss 稿同样:概括何时用、怎么接、怎么排错;命令模板、钩子名、JSON 字段、版本矩阵一律以 ZIP 内 SKILL.md 与 ClawHub 上游为准。
命令示例(摘自包内 SKILL.md)
以下为从上游 SKILL.md(或入库正文)自动抽取的终端/脚本片段;路径、环境变量与参数以当前 ZIP 与官方说明为准。
ClawHub slug:jwt(安装命令以 SKILL.md / claw CLI 为准)。
站内入库时的触发命令(完整语义见 ZIP):
# 使用本技能时可在对话中引用或执行上述指令;完整参数与示例见下载包内 SKILL.md。
/jwt
最佳实践
- 先 SKILL.md 再猜参数;站内专题稿不替代 schema 与必填字段说明。
- 委派任务时写清验收标准(命令、文件路径、测试命令),减少来回追问。
- 长任务用文档推荐的回调 / 日志落盘代替高频轮询,省 Token 也省机器负载。
- 多技能同时启用时,注意钩子加载顺序与重复工具调用(以 SKILL.md 冲突说明为准)。
调试与排错
- 打开 stderr 与客户端日志;PTY/tmux 场景同时看面板最后几十行输出。
- 参数错误时对照 SKILL.md 中的 JSON/CLI 示例(引号、转义、工作目录)。
- 网络类失败:查代理、防火墙、MCP 传输方式(stdio / HTTP / SSE)。
速查
| 动作 | 说明 |
|------|------|
| 获取技能包 | ClawHub / 站内 ZIP,核对版本 |
| 权威步骤 | 优先阅读 ZIP 内 SKILL.md |
| 首次试跑 | 使用 SKILL.md 最小示例 |
| 验收 | 对照路径、测试命令或回调负载 |
常见故障
- 无输出或立即退出 → 工作目录错误、依赖未装、或 Claude Code 未登录;按 SKILL.md 自检清单执行。
- 权限被拒绝 → 检查沙箱路径、
--permission-mode与工具白名单。 - 与简介不符 → 以英文 SKILL 与上游仓库为准,站内稿仅作结构化导读。
## Quick Reference
| Topic | File |
|-------|------|
| Algorithm selection | `algorithms.md` |
| Token lifecycle | `lifecycle.md` |
| Validation checklist | `validation.md` |
| Common attacks | `attacks.md` |
## Security Fundamentals
- JWTs are signed, not encrypted—anyone can decode and read the payload; never store secrets in it
- Always verify signature before trusting claims—decode without verify is useless for auth
- The `alg: none` attack: reject tokens with algorithm "none"—some libraries accepted unsigned tokens
- Use strong secrets: HS256 needs 256+ bit key; short secrets are brute-forceable
## Algorithm Choice
- HS256 (HMAC): symmetric, same key signs and verifies—good for single service
- RS256 (RSA): asymmetric, private key signs, public verifies—good for distributed systems
- ES256 (ECDSA): smaller signatures than RSA, same security—preferred for size-sensitive cases
- Never let the token dictate algorithm—verify against expected algorithm server-side
## Required Claims
- `exp` (expiration): always set and verify—tokens without expiry live forever
- `iat` (issued at): when token was created—useful for invalidation policies
- `nbf` (not before): token not valid until this time—for scheduled access
- Clock skew: allow 30-60 seconds leeway when verifying time claims
## Audience & Issuer
- `iss` (issuer): who created the token—verify to prevent cross-service token theft
- `aud` (audience): intended recipient—API should reject tokens for other audiences
- `sub` (subject): who the token represents—typically user ID
- Token confusion attack: without aud/iss validation, token for Service A works on Service B
## Token Lifecycle
- Access tokens: short-lived (5-15 min)—limits damage if stolen
- Refresh tokens: longer-lived, stored securely—used only to get new access tokens
- Refresh token rotation: issue new refresh token on each use, invalidate old one
- Revocation is hard—JWTs are stateless; use short expiry + refresh, or maintain blacklist
## Storage
- httpOnly cookie: immune to XSS, but needs CSRF protection
- localStorage: vulnerable to XSS, but simpler for SPAs
- Memory only: most secure, but lost on page refresh
- Never store in URL parameters—visible in logs, history, referrer headers
## Validation Checklist
- Verify signature with correct algorithm (don't trust header's alg)
- Check `exp` is in future (with clock skew tolerance)
- Check `iat` is not unreasonably old (optional policy)
- Verify `iss` matches expected issuer
- Verify `aud` includes your service
- Check `nbf` if present
## Common Mistakes
- Storing sensitive data in payload—it's just base64, not encrypted
- Huge payloads—JWTs go in headers; many servers limit header size to 8KB
- No expiration—indefinite tokens are security nightmares
- Same secret across environments—dev tokens work in production
- Logging tokens—they're credentials; treat as passwords
## Key Rotation
- Use `kid` (key ID) claim to identify which key signed the token
- JWKS (JSON Web Key Set) endpoint for public key distribution
- Overlap period: accept old key while transitioning to new
- After rotation, old tokens still valid until they expire—plan accordingly
## Implementation
- Use established libraries—don't implement JWT parsing yourself
- Libraries: `jsonwebtoken` (Node), `PyJWT` (Python), `java-jwt` (Java), `golang-jwt` (Go)
- Middleware should reject invalid tokens early—before any business logic