关于
The architect-executor skill executes pre-written implementation plans by orchestrating fresh, specialized agents for each discrete task, enforcing TDD and selective code review. It sequentially processes numbered plan tasks using a team of single-job specialists that operate without memory of previous work. Use this skill after a plan is created by writing-plans to handle task-by-task implementation with smart escalation.
快速安装
Claude Code
推荐npx skills add PackmindHub/packmind -a claude-code/plugin add https://github.com/PackmindHub/packmindgit clone https://github.com/PackmindHub/packmind.git ~/.claude/skills/architect-executor在 Claude Code 中复制并粘贴此命令以安装该技能
GitHub 仓库
Frequently asked questions
What is the architect-executor skill?
architect-executor is a Claude Skill by PackmindHub. Skills package instructions and resources that Claude loads on demand, so Claude can perform architect-executor-related tasks without extra prompting.
How do I install architect-executor?
Use the install commands on this page: add architect-executor to Claude Code as a plugin, or clone its repository into your skills directory, then restart Claude so it picks up the skill.
What category does architect-executor belong to?
architect-executor is in the Design category, tagged design.
Is architect-executor free to use?
Yes. architect-executor is listed on AIMCP and free to install. It runs inside Claude, so no separate service account is required to use the skill itself.
相关推荐技能
该Skill用于当开发者提供完整实施计划时，以受控批次方式执行代码实现。它会先审阅计划并提出疑问，然后分批次执行任务（默认每批3个任务），并在批次间暂停等待审查。关键特性包括分批次执行、内置检查点和架构师审查机制，确保复杂系统实现的可控性。
该Skill可在完成任务、实现主要功能或合并代码前自动调度代码审查子代理，确保实现符合需求和计划。它支持通过指定git SHA范围进行精准的代码变更审查，帮助开发者在关键节点及时发现潜在问题。核心原则是"早审查、勤审查"，适用于开发流程的各个关键阶段。
这个Skill指导开发者如何将MCP服务器连接到Claude Code，支持HTTP、stdio和SSE三种传输协议。它涵盖了从安装配置到认证安全的完整流程，适用于集成GitHub、Notion、数据库等外部服务。当开发者需要添加集成、配置外部工具或提及MCP相关功能时，这个Skill能提供实用的操作指南。
该Skill帮助开发者根据任务特性选择Claude Code的Web或CLI界面，并指导如何在两种环境间无缝迁移会话。它能分析任务复杂度、迭代需求等要素，推荐最优工作界面和工作流。关键特性包括会话状态管理、环境切换指导和上下文优化建议。