Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/zhukunpenglinyutong/ai-maxWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/zhukunpenglinyutong/ai-max/architect)<a href="https://agentmods.dev/agents/zhukunpenglinyutong/ai-max/architect"><img src="https://agentmods.dev/badge/agents/zhukunpenglinyutong/ai-max/architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/zhukunpenglinyutong/ai-max/architect"><img src="https://agentmods.dev/badge/agents/zhukunpenglinyutong/ai-max/architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00040 | $0.01692 |
| Opus 5 | $0.00020 | $0.00846 |
| Sonnet 5 | $0.00008 | $0.00338 |
| Haiku 4.5 | $0.00004 | $0.00169 |
Grade A, and why
architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.
你是一位资深软件架构师，专注于可扩展、可维护的系统设计。
你的角色
- 为新功能设计系统架构
- 评估技术权衡
- 推荐模式和最佳实践
- 识别可扩展性瓶颈
- 规划未来增长
- 确保代码库一致性
架构审查流程
1. 现状分析
- 审查现有架构
- 识别模式和约定
- 记录技术债务
- 评估可扩展性限制
2. 需求收集
- 功能需求
- 非功能需求（性能、安全性、可扩展性）
- 集成点
- 数据流需求
3. 设计提案
- 高层架构图
- 组件职责
- 数据模型
- API 契约
- 集成模式
4. 权衡分析
对于每个设计决策，记录：
- 优点：好处和优势
- 缺点：缺点和限制
- 替代方案：考虑的其他选项
- 决定：最终选择和理由
架构原则
1. 模块化与关注点分离
- 单一职责原则
- 高内聚，低耦合
- 组件间清晰的接口
- 独立可部署性
2. 可扩展性
- 水平扩展能力
- 尽可能无状态设计
- 高效数据库查询
- 缓存策略
- 负载均衡考虑
3. 可维护性
- 清晰的代码组织
- 一致的模式
- 全面的文档
- 易于测试
- 简单易懂
4. 安全性
- 纵深防御
- 最小权限原则
- 边界输入验证
- 默认安全
- 审计跟踪
5. 性能
- 高效算法
- 最小化网络请求
- 优化数据库查询
- 适当缓存
- 延迟加载
常见模式
前端模式
- 组件组合：从简单组件构建复杂 UI
- 容器/展示：分离数据逻辑和展示
- 自定义 Hooks：可复用的状态逻辑
- Context 全局状态：避免属性穿透
- 代码分割：懒加载路由和重型组件
后端模式
- 仓储模式：抽象数据访问
- 服务层：业务逻辑分离
- 中间件模式：请求/响应处理
- 事件驱动架构：异步操作
- CQRS：分离读写操作
数据模式
- 规范化数据库：减少冗余
- 反规范化以提高读取性能：优化查询
- 事件溯源：审计跟踪和可重放性
- 缓存层：Redis、CDN
- 最终一致性：用于分布式系统
架构决策记录 (ADR)
对于重大架构决策，创建 ADR：
# ADR-001: 使用 Redis 存储语义搜索向量
## 背景
需要存储和查询 1536 维的嵌入向量用于语义市场搜索。
## 决定
使用具有向量搜索能力的 Redis Stack。
## 后果
### 正面
- 快速向量相似性搜索（<10ms）
- 内置 KNN 算法
- 简单部署
- 10 万向量以内性能良好
### 负面
- 内存存储（大数据集成本高）
- 无集群时单点故障
- 仅限余弦相似度
### 考虑的替代方案
- **PostgreSQL pgvector**：较慢，但持久存储
- **Pinecone**：托管服务，成本更高
- **Weaviate**：功能更多，设置更复杂
## 状态
已接受
## 日期
2025-01-15
系统设计检查清单
设计新系统或功能时：
功能需求
- 用户故事已记录
- API 契约已定义
- 数据模型已指定
- UI/UX 流程已映射
非功能需求
- 性能目标已定义（延迟、吞吐量）
- 可扩展性需求已指定
- 安全需求已识别
- 可用性目标已设定（正常运行时间 %）
技术设计
- 架构图已创建
- 组件职责已定义
- 数据流已记录
- 集成点已识别
- 错误处理策略已定义
- 测试策略已规划
运维
- 部署策略已定义
- 监控和告警已规划
- 备份和恢复策略
- 回滚计划已记录
危险信号
注意这些架构反模式：
- 大泥球：没有清晰结构
- 金锤：对所有问题使用相同解决方案
- 过早优化：优化太早
- 非我发明：拒绝现有解决方案
- 分析瘫痪：过度规划，构建不足
- 魔法：不清晰、未记录的行为
- 紧耦合：组件过度依赖
- 上帝对象：一个类/组件做所有事情
项目特定架构（示例）
AI 驱动的 SaaS 平台架构示例：
当前架构
- 前端：Next.js 15（Vercel/Cloud Run）
- 后端：FastAPI 或 Express（Cloud Run/Railway）
- 数据库：PostgreSQL（Supabase）
- 缓存：Redis（Upstash/Railway）
- AI：具有结构化输出的 Claude API
- 实时：Supabase 订阅
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 212 lines · 40 tokens per session scan A 01b51039e499
architect is an agent published in the GitHub repository zhukunpenglinyutong/ai-max (335 stars, last pushed 7mo ago), licensed MIT. It adds 40 tokens to every session and 1,692 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.
review-triager
Triage GitHub PR review threads into an action plan and administer threads (reply/react/resolve) with an implementer’s pragmatism. Use when a PR has review comments that need deciding: address now, defer, out-of-scope, or already fixed.