architect-mode SKILL.md skillA

architect-mode from chainlesschain/chainlesschain is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add chainlesschain/chainlesschain/architect-mode@v1.0.0

Pins v1.0.0 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
chainlesschain/chainlesschain
Path
desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/architect-mode/SKILL.md
Versions recorded
3
Latest version
v1.0.0, committed Aug 28, 2026
First version recorded
Jun 3, 2026
Size
8,286 bytes
Repository stars
11
License
a license file GitHub does not identify

双阶段架构模式 - 先规划后编辑,分离推理与代码修改,提升多文件编辑准确率.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Architect Mode 技能

## 描述

双阶段架构模式,将复杂的多文件编辑任务拆分为"规划"与"执行"两个独立阶段。通过分离推理(reasoning)和代码修改(editing),显著提升多文件变更的准确率和可控性。灵感来源于 Aider 的 architect mode 和 Claude Code 的 plan mode。

## 核心理念

传统 AI 编码助手在处理多文件变更时容易出现:

- 遗漏需要修改的文件
- 修改顺序导致中间状态不一致
- "迷失在上下文中"忘记原始目标

Architect Mode 解决这些问题的方式:

1. **Phase 1 (规划)**: 使用推理能力强的模型分析代码库,生成结构化变更计划
2. **Phase 2 (执行)**: 按照计划逐个文件生成精确的编辑指令

## 使用方法

```
/architect-mode <任务描述>              # 完整流程:规划 → 审核 → 执行
/architect-mode --plan-only <任务描述>  # 仅生成计划,不执行
/architect-mode --execute <plan-id>     # 执行已有计划
```

## 操作模式

### 默认模式(规划 + 执行)

```
/architect-mode 'add WebSocket support for real-time notifications'
```

流程:

1. 扫描代码库,提取与任务相关的文件
2. 生成结构化变更计划(分阶段、按依赖排序)
3. 输出计划摘要,等待用户审核
4. 用户确认后,逐个文件生成编辑指令
5. 输出执行结果和摘要

### 仅规划模式

```
/architect-mode --plan-only 'migrate from Express to Fastify'
```

只生成变更计划,不自动执行。适用于:

- 大型重构前的影响评估
- 团队讨论变更方案
- 多方案对比

### 执行已有计划

```
/architect-mode --execute plan-abc123
```

执行之前生成并保存的计划。支持断点续传。

## 计划结构

```json
{
  "id": "plan-abc123",
  "task": "add user authentication with JWT",
  "createdAt": "2026-02-17T10:30:00Z",
  "phases": [
    {
      "order": 1,
      "description": "Create authentication infrastructure",
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add chainlesschain/chainlesschain/architect-mode@sha256:38d567508e2279bd

The line at the top of this page pins the label v1.0.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_65agyv27ic5u76fn.svg)](https://markdownregistry.com/a/art_65agyv27ic5u76fn)

5 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 3 versions of this file, between Jun 3, 2026 and Aug 28, 2026, about one every 43 days.

versioncommittedcommitsizeaudit
v1.0.0 latest2026-08-28 c307200 8,286 BA view · diff
v1.0.02026-08-28 c29953e 8,269 BA view · diff
v1.0.02026-06-03 3b0b2dc 8,129 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (8286 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

chainlesschain/chainlesschain · 11 stars · license a license file GitHub does not identify · pushed 2026-11-02 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_65agyv27ic5u76fn
GET https://markdownregistry.com/api/v1/resolve?ref=chainlesschain/chainlesschain/architect-mode
GET https://markdownregistry.com/api/v1/blob/38d567508e2279bd2e355261583496da4729180add1a5fbd23a5eca1dd910a03

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from chainlesschain/chainlesschain

cc-cli SKILL.md skill
chainlesschain/chainlesschain · .claude/skills/cc-cli/SKILL.md · Invoke ChainlessChain `cc` CLI commands from natural language. Triggers when the user wants to run any `cc …` action…
git:20260618.4898c75 · audit A · 11 stars
pdh-android-collect SKILL.md skill
chainlesschain/chainlesschain · .claude/skills/pdh-android-collect/SKILL.md · Collect a rooted Android device's real personal data into the Personal Data Hub (PDH) vault to feed the user's…
git:20260618.58d3bd9 · audit A · 11 stars
AGENTS.md AGENTS.md file
chainlesschain/chainlesschain · AGENTS.md
git:20261023.f06ee98 · audit A · 11 stars
CLAUDE.md CLAUDE.md file
chainlesschain/chainlesschain · CLAUDE.md
git:20260920.96ab90a · audit A · 11 stars
ab-compare SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/ab-compare/SKILL.md · Generate and compare multiple implementation variants with benchmarking and scoring
v1.0.0 · audit A · 11 stars
agent-browser SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/agent-browser/SKILL.md · Advanced browser automation for AI agents with snapshot-ref interaction pattern - navigate, snapshot interactive…
v1.0.0 · audit A · 11 stars
api-design SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/api-design/SKILL.md · Design RESTful APIs with best practices, conventions, and OpenAPI specs
v1.0.0 · audit A · 11 stars
api-docs-generator SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/api-docs-generator/SKILL.md · Generate API documentation from code - produce OpenAPI/Swagger specs, Markdown API references, request/response…
v1.0.0 · audit A · 11 stars
api-gateway SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/api-gateway/SKILL.md · Universal API gateway - connect to 100+ APIs with a unified interface, manage API keys, and chain API calls
v1.2.0 · audit A · 11 stars
api-tester SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/api-tester/SKILL.md · API测试技能 - IPC处理器发现、测试生成、健康检查、回归测试
v1.0.0 · audit A · 11 stars
audio-transcriber SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/audio-transcriber/SKILL.md · Speech-to-text transcription using Whisper API or local engine
v1.0.0 · audit A · 11 stars
auto-context SKILL.md skill
chainlesschain/chainlesschain · desktop-app-vue/src/main/ai-engine/cowork/skills/builtin/auto-context/SKILL.md · 智能上下文检测 - 自动识别AI所需的文件、函数和上下文,优化token使用
v1.0.0 · audit A · 11 stars

Every file in chainlesschain/chainlesschain

Browse by kind, by grade A, or by owner.