deepinit 生成的子目录级 AGENTS.md 文件,帮助 AI agents 理解各功能模块: - docs/config/: advanced, basic, caching, load-balancing, lua, rewriting, security, ssl - docs/lua/: api-gateway, authentication, caching, dynamic-routing, logging-monitoring, middleware, rate-limiting, websocket Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
44 lines
1.4 KiB
Markdown
44 lines
1.4 KiB
Markdown
<!-- Parent: ../AGENTS.md -->
|
||
<!-- Generated: 2026-04-23 | Updated: 2026-04-23 -->
|
||
|
||
# security
|
||
|
||
## Purpose
|
||
安全配置示例目录,包含访问控制、认证、限流、连接限制、安全头部等安全防护配置。
|
||
|
||
## Key Files
|
||
|
||
| File | Description |
|
||
|------|-------------|
|
||
| `access-control.conf` | 访问控制:IP 白名单/黑名单、allow/deny 规则 |
|
||
| `basic-auth.conf` | Basic 认证:用户名密码保护、htpasswd 文件 |
|
||
| `auth-request.conf` | 子请求认证:外部认证服务集成 |
|
||
| `rate-limit.conf` | 请求限流:令牌桶、漏桶、请求速率控制 |
|
||
| `conn-limit.conf` | 连接限制:并发连接数、单 IP 连接限制 |
|
||
| `security-headers.conf` | 安全头部:X-Frame-Options、CSP、HSTS |
|
||
|
||
## For AI Agents
|
||
|
||
### Working In This Directory
|
||
- 访问控制按顺序匹配:先检查黑名单,再检查白名单
|
||
- Basic Auth 使用 bcrypt 密码哈希
|
||
- 限流使用令牌桶算法,支持按 IP/全局限流
|
||
- 安全头部建议全部启用
|
||
|
||
### Testing Requirements
|
||
- 安全配置通过单元测试验证
|
||
- 运行测试:`go test ./internal/middleware/security/...`
|
||
|
||
### Common Patterns
|
||
- 认证失败返回 401 Unauthorized
|
||
- 限流触发返回 429 Too Many Requests
|
||
- 访问拒绝返回 403 Forbidden
|
||
|
||
## Dependencies
|
||
|
||
### Internal
|
||
- `../../../internal/middleware/security/` - 安全中间件实现
|
||
- `../../../internal/config/` - 安全配置解析
|
||
|
||
<!-- MANUAL: -->
|