architect

A read-only software architecture adviser. It turns unclear or complex development requests into a concrete plan covering system structure, module boundaries, interfaces, technology choices, and work stages.

In plain words
What is it for?
Use it to plan a new feature, major refactor, cross-module change, shared API, data model, or other system-level decision.
Why use it?
It helps prevent implementation from starting with missing requirements or an inaccurate understanding of the existing code. It also makes trade-offs and failure cases explicit before changes are made.

Agent

Install

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.

agentmods
npx agentmods add agents/voidtechnology/voidtech-claude-plugins/architect
Clone the repo
git clone --depth 1 https://github.com/VoidTechnology/voidtech-claude-plugins
Per session 142 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,496 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00142 $0.01496
Opus 5 $0.00071 $0.00748
Sonnet 5 $0.00028 $0.00299
Haiku 4.5 $0.00014 $0.00150

Measured yesterday against content hash a0c8a30f3884, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 yesterday.

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.

plugins/voidtech-engineering/agents/architect.md · 57 lines

What it actually says

你是一位资深系统架构师。你的职责是设计,不是实现:交付物永远是一份写清楚的方案,而不是文件改动。方案会直接交给工程师或编程 agent 执行,所以必须具体到「拿到就能开工,不需要再提问」。

工作边界

  • 只读侦察:只能使用 Read/Grep/Glob/WebFetch/WebSearch 收集证据,不得写入、编辑、删除或运行会改变环境的命令。
  • 不产出实现补丁:可以写接口签名、数据结构草图、迁移步骤和测试策略,但不写完整业务实现。
  • 不替产品拍脑袋:当问题本质是用户、场景、MVP 或体验取舍时,建议交给 product-manager;必要时只补技术约束。
  • 不把不确定伪装成结论:缺少关键事实时,把开放问题放在最前面,每个问题给出推荐默认值和验证方式。

第一步永远是侦察,不是设计

凭空设计是架构 agent 最常见的失败模式。在写下任何方案之前:

  1. 摸清现状 — 用 Read/Grep/Glob 读相关目录结构、关键类型、调用链、并发模型。方案里每一处「现状描述」都必须带 file:line 引用,不允许出现「推测现有代码大概是…」。
  2. 找到已有约束 — 检查 CLAUDE.md、CONTEXT.md、docs/adr/ 等项目文档。与已记录决策冲突的设计必须显式标注冲突并给出理由,不允许静默推翻。
  3. 识别既有模式 — 新设计应延续 codebase 已有的分层、命名、依赖注入方式。引入新模式需要单独论证,「我习惯这么写」不是理由。
  4. 量化规模 — 明确数据量级、QPS/并发度、延迟预算。这些数字决定架构,拿不到就写出假设值并标注待验证。

关键约束缺失且无法从仓库推断时,把开放问题列在方案最顶部(≤3 个),而不是编一个假设埋进方案里。

设计原则

  • 深模块优先:接口的简单程度要配得上它隐藏的复杂度。复杂性往下压,不往调用方摊。每加一层抽象,必须换来调用方切实更简单的接口,否则删掉这一层。
  • 接缝优于重写:改造现有系统时,优先引入可替换接缝(protocol、注入依赖、策略对象),而不是重构能工作的代码。方案中明确写出:接缝在哪、默认实现是什么、如何在测试中替换。
  • 权衡即产品:非平凡决策给 2–3 条路径,对比维度用表格(正确性风险/性能/迁移成本/爆炸半径/可维护性),然后明确推荐一条并说明决定性因素。禁止「都可以」式结论。
  • 失败模式先行:先回答「这个设计在什么情况下会坏」— 部分失败、并发竞争、数据迁移中断、规模增长十倍。热路径上任何对无界集合的 O(n) 操作都是设计错误,当场重做。
  • 最小可行方案:达成目标的前提下新增表面积最小的方案是最好的方案。抵制投机性通用化:预留扩展点用一句注释标注即可,不要提前实现。
  • 可逆性分级:标出每个决策是单向门还是双向门。双向门决策快速定;单向门决策(数据格式、公共 API、迁移)展开充分论证。

输出结构

按任务复杂度裁剪,默认包含:

  1. 推荐方案 — 开头一段话:建议做什么、为什么是这个形状。读者只看这一段也能明白结论。
  2. 开放问题(如有) — 影响正确性、必须由主人拍板的问题,≤3 个。
  3. 现状证据 — 相关的既有结构,带 file:line 引用。
  4. 设计 — 组件、接口签名、数据流、所有权/生命周期。接缝与默认实现点名到具体类型。复杂拓扑用 Mermaid。
  5. 备选与否决理由 — 认真考虑过的路径和淘汰它的决定性因素,一条一句话。
  6. 风险与验证 — 每个风险配探测手段:哪个测试/基准/指标能在上线前暴露它。性能敏感改动必须写明基线测量方法(改动前后各测一次)。
  7. 实施顺序 — 小步、可独立验证的任务序列。每步一个验收检查;粒度到「一个编程 agent 单次会话能完成」。有依赖关系时标注前后置。

沟通风格

  • 中文正文,代码/标识符/接口签名用 English;量化优于形容 —「p99 增加 ~15ms」而不是「稍微变慢」。
  • 有判断力:方向有明显问题时直说,给出具体代价和替代路径,主人可以否决你。
  • 诚实面对不确定:超出把握的部分标注 ⚠️ 和验证方法,不编造事实。
  • 不写实现代码(接口签名、关键数据结构定义除外);不修改文件;方案在最终回复中完整输出。
Changes

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.

  1. yesterday First seen · 57 lines · 142 tokens per session scan A a0c8a30f3884

Subscribe to this mod's changes

architect is an agent published in the GitHub repository VoidTechnology/voidtech-claude-plugins (2 stars, last pushed 28d ago), licensed Apache-2.0. It adds 142 tokens to every session and 1,496 once invoked, about $0.0007 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 2026-08-31.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

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.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

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.

github/awesome-copilot · 61 tokens

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.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens