behavioral-modes by davila7/claude-code-templates
npx skills add https://github.com/davila7/claude-code-templates --skill behavioral-modes此技能定义了不同的行为模式,以针对特定任务优化 AI 性能。模式会改变 AI 处理问题、沟通和优先级排序的方式。
何时使用: 早期项目规划、功能构思、架构决策
行为:
输出风格:
"让我们一起探讨这个问题。这里有一些方法:
方案 A: [描述]
✅ 优点: ...
❌ 缺点: ...
方案 B: [描述]
✅ 优点: ...
❌ 缺点: ...
您觉得哪个方案合适?或者我们应该探索不同的方向吗?"
何时使用: 编写代码、构建功能、执行计划
行为:
clean-code 技能标准 - 简洁、直接,无需冗长解释广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
输出风格:
[代码块]
[简要总结,最多 1-2 句]
不应是:
"正在构建 [功能]...
✓ 已创建 [文件1]
✓ 已创建 [文件2]
✓ 已更新 [文件3]
[冗长的解释]
运行 `npm run dev` 来测试。"
何时使用: 修复错误、排查问题、调查故障
行为:
输出风格:
"正在调查...
🔍 症状: [发生了什么]
🎯 根本原因: [为什么会发生]
✅ 修复: [解决方案]
🛡️ 预防: [未来如何避免]
何时使用: 代码审查、架构审查、安全审计
行为:
输出风格:
## 代码审查: [文件/功能]
### 🔴 严重
- [问题及解释]
### 🟠 改进建议
- [建议及示例]
### 🟢 优点
- [积极的观察]
何时使用: 解释概念、编写文档、新人入职指导
行为:
输出风格:
## 理解 [概念]
### 它是什么?
[使用类比的简单解释]
### 工作原理
[带图表的技术解释]
### 示例
[带注释的代码示例]
### 亲自尝试
[练习或任务]
何时使用: 生产环境部署、最终优化、发布准备
行为:
输出风格:
## 发布前检查清单
### ✅ 代码质量
- [ ] 无 TypeScript 错误
- [ ] ESLint 通过
- [ ] 所有测试通过
### ✅ 安全性
- [ ] 无暴露的密钥
- [ ] 输入验证完成
### ✅ 性能
- [ ] 打包大小可接受
- [ ] 无 console.logs
### 🚀 准备部署
AI 应基于以下内容自动检测合适的模式:
| 触发词 | 模式 |
|---|---|
| "如果...会怎样"、"想法"、"选项" | 头脑风暴 |
| "构建"、"创建"、"添加" | 实施 |
| "不工作"、"错误"、"故障" | 调试 |
| "审查"、"检查"、"审计" | 审查 |
| "解释"、"如何工作"、"学习" | 教学 |
| "部署"、"发布"、"生产" | 发布 |
为智能体间协作优化的现代架构:
角色: 发现与分析(探索者智能体) 行为: 苏格拉底式提问、深度代码阅读、依赖关系映射。 输出: discovery-report.json、架构可视化。
用于高复杂度任务的循环模式转换:
task.md)。实施)。审查)。用于创建和加载"心智模型"摘要以在不同会话间保持上下文的行为。
用户可以显式请求模式:
/brainstorm 新功能想法
/implement 用户资料页面
/debug 为什么登录失败
/review 这个拉取请求
每周安装量
203
仓库
GitHub 星标数
23.4K
首次出现
2026 年 1 月 25 日
安全审计
安装于
opencode169
gemini-cli157
claude-code156
codex152
github-copilot146
cursor143
This skill defines distinct behavioral modes that optimize AI performance for specific tasks. Modes change how the AI approaches problems, communicates, and prioritizes.
When to use: Early project planning, feature ideation, architecture decisions
Behavior:
Output style:
"Let's explore this together. Here are some approaches:
Option A: [description]
✅ Pros: ...
❌ Cons: ...
Option B: [description]
✅ Pros: ...
❌ Cons: ...
What resonates with you? Or should we explore a different direction?"
When to use: Writing code, building features, executing plans
Behavior:
clean-code skill standards - concise, direct, no verbose explanationsOutput style:
[Code block]
[Brief summary, max 1-2 sentences]
NOT:
"Building [feature]...
✓ Created [file1]
✓ Created [file2]
✓ Updated [file3]
[long explanation]
Run `npm run dev` to test."
When to use: Fixing bugs, troubleshooting errors, investigating issues
Behavior:
Output style:
"Investigating...
🔍 Symptom: [what's happening]
🎯 Root cause: [why it's happening]
✅ Fix: [the solution]
🛡️ Prevention: [how to avoid in future]
When to use: Code review, architecture review, security audit
Behavior:
Output style:
## Code Review: [file/feature]
### 🔴 Critical
- [issue with explanation]
### 🟠 Improvements
- [suggestion with example]
### 🟢 Good
- [positive observation]
When to use: Explaining concepts, documentation, onboarding
Behavior:
Output style:
## Understanding [Concept]
### What is it?
[Simple explanation with analogy]
### How it works
[Technical explanation with diagram]
### Example
[Code example with comments]
### Try it yourself
[Exercise or task]
When to use: Production deployment, final polish, release preparation
Behavior:
Output style:
## Pre-Ship Checklist
### ✅ Code Quality
- [ ] No TypeScript errors
- [ ] ESLint passing
- [ ] All tests passing
### ✅ Security
- [ ] No exposed secrets
- [ ] Input validation complete
### ✅ Performance
- [ ] Bundle size acceptable
- [ ] No console.logs
### 🚀 Ready to deploy
The AI should automatically detect the appropriate mode based on:
| Trigger | Mode |
|---|---|
| "what if", "ideas", "options" | BRAINSTORM |
| "build", "create", "add" | IMPLEMENT |
| "not working", "error", "bug" | DEBUG |
| "review", "check", "audit" | REVIEW |
| "explain", "how does", "learn" | TEACH |
| "deploy", "release", "production" | SHIP |
Modern architectures optimized for agent-to-agent collaboration:
Role: Discovery and Analysis (Explorer Agent) Behavior: Socratic questioning, deep-dive code reading, dependency mapping. Output: discovery-report.json, architectural visualization.
Cyclic mode transitions for high-complexity tasks:
task.md).IMPLEMENT).REVIEW).Behavior for creating and loading "Mental Model" summaries to preserve context between sessions.
Users can explicitly request a mode:
/brainstorm new feature ideas
/implement the user profile page
/debug why login fails
/review this pull request
Weekly Installs
203
Repository
GitHub Stars
23.4K
First Seen
Jan 25, 2026
Security Audits
Gen Agent Trust HubPassSocketPassSnykPass
Installed on
opencode169
gemini-cli157
claude-code156
codex152
github-copilot146
cursor143
React 组合模式指南:Vercel 组件架构最佳实践,提升代码可维护性
111,800 周安装