aios-arch

aios-arch is a skill for Claude Code, Codex from ArchSightLabs/archsight-aios. It costs 57 tokens per session (4,336 once invoked), scanned A, original, Apache-2.0.

A workflow for reviewing software architecture, including service boundaries, technical choices, data and model boundaries, AI workflows, reliability, and long-term complexity.

In plain words
What is it for?
Use it to review backend systems, platforms, GraphRAG systems, agent workflows, file processing, queues, indexes, caches, permissions, audit logs, and recovery paths.
Why use it?
It helps identify design trade-offs, failure risks, unclear ownership, and migration costs before they become harder to change.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present. Also seen: mentions Codex.

Part of the archsight-aios plugin — 33 skills shipped together

Good fit Use it to review backend systems, platforms, GraphRAG systems, agent workflows, file processing, queues, indexes, caches, permissions, audit logs, and recovery paths.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/archsightlabs/archsight-aios/aios-arch
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.

Any agent
npx skills add ArchSightLabs/archsight-aios --skill aios-arch
Clone the repo
git clone --depth 1 https://github.com/ArchSightLabs/archsight-aios

Made for: Claude Code, Codex.

Or install archsight-aios, the plugin that ships this one along with the rest of its 33 skills.

Wrote 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.

agentmods badge for aios-arch

README.md
[![agentmods](https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-arch/github.svg)](https://agentmods.dev/skills/archsightlabs/archsight-aios/aios-arch)
Your own site
<a href="https://agentmods.dev/skills/archsightlabs/archsight-aios/aios-arch"><img src="https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-arch/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.

agentmods 80×15 button for aios-arch

Your own site · 80×15
<a href="https://agentmods.dev/skills/archsightlabs/archsight-aios/aios-arch"><img src="https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-arch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,336 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 5.1 $0.00057 $0.04336
Opus 5 $0.00028 $0.02168
Sonnet 5 $0.00011 $0.00867
Haiku 4.5 $0.00006 $0.00434

Measured 11d ago against content hash 4288db1a6b13, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

aios-arch 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 11d 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.

skills/aios-arch/SKILL.md · 236 lines

How it starts

The opening of the file, as written. The whole thing — 236 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AIOS Arch

目标

以 Atlas(总架构师)的方式审查方案:先判断边界和长期复杂度,再给出可落地的推荐路径。适用于 Codex、Gemini 或其他 AI 编程助手在项目工作目录中执行架构评审。

AIOS Arch 的目标是补足通用架构评审:在 AIOS 行业增强启用时,把行业语义、工程证据链、审计可追溯性和后端运行可靠性纳入默认检查。

没有 .ai/ 目录也可以使用本 Skill。此时优先读取代码、接口、schema、配置、测试、部署入口和用户提供的行业背景;只有任务事实明确涉及建筑行业语义时,才引入 BIM、IFC、规范知识或审图假设。

AIOS 适用性

本 Skill 继承 AIOS 的全局定位:AIOS 是建筑行业增强层,不是通用任务替代器。

  • 建筑行业项目、平台、系统、数据链路或 AI Runtime 的架构评审,启用 AIOS 行业增强。
  • 普通非建筑架构问题优先使用宿主工具的通用架构能力;不要为了“已安装 AIOS”强行套用 BIM、IFC、规范、审图或工程证据链假设。
  • 是否适用不明确时,先读 README、.ai/project-context.md、项目 profile 和当前任务事实。

适用对象

  • 建筑行业架构师:关注平台边界、模型 / 图纸 / 规范 / 审图链路、可审计性和长期演进成本。
  • 博士 / 研究型团队:关注算法假设、RAG / GraphRAG 方案、评估集、实验可复现和工程落地边界。
  • 后端开发:关注服务边界、任务队列、文件处理、索引版本、缓存、多实例、权限、审计日志和失败恢复。

与 CEO / Product 的组合

  • aios-ceo + aios-arch 是一等联合评审:两者共用事实底稿,但分别回答战略可信度与技术可信度,最后只合并一致项、冲突项和处理建议。
  • aios-product <-> aios-arch 是产品契约与技术边界的迭代:Atlas 对需求逐项返回 支持 / 需调整 / 技术阻断,说明项目事实、失败模式和验证路径;不自行重排用户价值优先级。
  • 三者同时使用时,CEO 决定目标用户、价值、投入和停损边界;Product 定义该边界内的版本范围、非目标和 UAT;Arch 判断技术边界、可靠性、迁移代价与可验证性。
  • 技术约束只改变实现方式时直接反馈 Product;需要牺牲核心用户结果、改变目标市场、扩大投入或触发停损线时升级 CEO。
  • 用户明确要求 CEO + Arch 时,不因出现功能清单自动强制调用 Product;只有需要进入具体版本契约、PRD 或试点闭环时才交接。

输入

优先收集最小必要上下文:

  • 需求背景和当前问题。
  • 已确认的 CEO 阶段决策或 Product 版本契约,如存在;没有时明确当前架构判断依赖哪些产品假设。
  • 相关目录、模块、接口或数据结构。
  • 现有代码、配置、契约、测试、脚本、部署入口和运行方式。
  • 已有设计方案或候选方案。
  • 约束条件:时间、成本、团队、技术栈、数据、权限、运行环境。
  • 已知风险、测试结果或失败记录。
  • 可用 Capability、工具返回值、规范查询、结构求解、测试 / 构建 / 安全扫描证据。

信息不足时,先列出缺口和可推进的最小判断,不要编造背景。

工作流

  1. 明确问题类型:平台边界、服务边界、数据模型、技术选型、Runtime、RAG / GraphRAG、Agent 协同或长期演进。
  2. 读取项目约定和相关代码事实;文档只能作为输入之一,必须尽量用代码、契约、测试、配置或部署入口核验。
  3. 做 Step 0 技术范围挑战:先判断当前技术方案是否值得进入架构评审,避免在错误实现范围里做深度优化;不替代 CEO 判断项目是否值得做。
  4. 盘点已有能力:列出可复用的模块、契约、测试、脚本和治理资产,优先说明“不需要重建什么”。
  5. 抽样追踪关键端到端链路:选择至少一个用户输入、配置字段、领域元数据、版本关系、审计关系或跨存储关系,从入口追到领域模型、任务、存储、消费端和测试。
  6. 按工程评审维度逐项审查:架构、实现质量、测试 / eval、性能 / 可运维性。
  7. 判断现有方案是否最小、稳定、可验证。
  8. 识别耦合点、复杂度来源、技术债、生产失效方式和后续迁移成本。
  9. 用 P0/P1/P2 或同等级别标注风险优先级;不要把所有问题写成平级 TODO。
  10. 做交付审查增强:输出事实刷新、历史结论 diff、领域风险 / 工程风险分类、任务化落点和第一步建议。
  11. 给出推荐方案,并说明被拒绝方案和原因。
  12. 对多 Agent 冲突输出中文化的 判断事项 / 证据 / 工具结果 / 处理建议,按 governance/arbitration-protocol.md 仲裁。
  13. 给 Janus、Mason、Daedalus、Argus、Vitruvius、Euclid 或 Hephaestus 标注后续交接点;用户问题、版本范围和产品优先级交回 aios-product,工程拆解细节交给 Mason,不在 Atlas 报告里替代产品或交付计划。

Read the full file on GitHub · 236 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 11d ago First seen · 236 lines · 57 tokens per session scan A 4288db1a6b13

Subscribe to this mod's changes

aios-arch is a skill published in the GitHub repository ArchSightLabs/archsight-aios (15 stars, last pushed 14d ago), licensed Apache-2.0. It adds 57 tokens to every session and 4,336 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

massing-bim

Drive a Massing BIM/AEC project from an AI agent over MCP — read a project's status, records, CDE, KPI and model-quality checks; run standards-compliance, schedule-risk, embodied-carbon, permit- readiness and drawing-QA analyses; author the IFC model with GUID-stable recipes; and draft RFIs. Use when the user asks to…

ibuilder/massing · 90 tokens

ara3d-sdk

Use when processing AEC/BIM 3D data in .NET 8 — mesh generation and transformation, SIMD-accelerated math, IFC/STEP/PLY to glTF/GLB/VIM conversion, plugin development. Ara3D-SDK: high-performance .NET 3D geometry and BIM library suite.

znlgis/opengis-skills · 70 tokens

elements

Use when generating building information models (BIM) programmatically in C#/.NET — wall, beam, column, floor creation, geometry kernel (BREP/CSG), glTF/IFC/JSON serialization, MEP systems, spatial grids. Hypar Elements: the smallest useful BIM — a cross-platform C# library for creating building elements without…

znlgis/opengis-skills · 81 tokens

3d-skills

Use when working with 3D Gaussian Splatting (3DGS), .ply model cleanup/compression/publishing, interactive 360-degree panorama visualization and virtual tours, AEC/BIM 3D data processing, or CSG solid modeling. Index of 5 skills: SuperSplat, Photo-Sphere-Viewer, Ara3D-SDK, Elements, and OpenCSG.NET.

znlgis/opengis-skills · 86 tokens

naviscoord-setup

Instalar, actualizar, reparar y verificar NavisCoord en ChatGPT Work de escritorio, Claude Desktop, Claude Code y Codex cuando faltan las herramientas navis o el puente de Navisworks no responde.

HorizunGroup/naviscoord-mcp · 50 tokens

bim-agent

BIM Agent — Query and validate Building Information Models (IFC) through natural language.

Daviidro/bim-agent · 21 tokens