Leadership Prompts
Battle-tested prompt library for engineering managers, tech leads, VPs of Engineering, and CTOs. These aren't generic "tell me about leadership" prompts — they're specific, opinionated frameworks for real challenges you face weekly.
Who This Is For
- - Engineering Managers running teams of 5-15
- Tech Leads navigating the IC-to-management boundary
- Directors/VPs managing managers
- CTOs who still get their hands dirty
Categories
| Category | Count | When to reach for it |
|---|
| 1-on-1 Prep | 4 | Before any non-routine 1-on-1 |
| Team Health |
4 | After rough quarters, conflicts, layoffs, or remote disconnection |
|
Incident Retrospectives | 3 | Within 48h of incident resolution |
|
Technical Strategy | 4 | Quarterly planning, architecture reviews, build-vs-buy decisions |
|
Hiring & Interviews | 3 | Opening new roles, optimizing pipeline, closing candidates |
|
Career Development | 4 | Promotion cycles, feedback delivery, retention conversations |
|
Stakeholder Communication | 4 | Exec updates, saying no, cross-functional alignment, reorgs |
Quick Start
Using the CLI
CODEBLOCK0
Using with Your AI Assistant
Just tell your assistant:
"I need to prepare for a 1-on-1 with an underperformer. Use the leadership-prompts skill to find the right prompt, then walk me through it."
Or for browsing:
"Show me all the hiring prompts from leadership-prompts"
How to Use a Prompt
Each prompt has placeholder variables in {curly_braces}. Fill these in with your specific context. The more specific you are, the better the output.
Example:
The prompt says:
I'm preparing for a 1-on-1 with a direct report who has been underperforming for the past {timeframe}...
You fill in:
I'm preparing for a 1-on-1 with a direct report who has been underperforming for the past 6 weeks...
Prompt Design Principles
These prompts are designed to:
- 1. Force structure — They use frameworks (SBI, RACI, RAG status) so your output is actionable, not rambling
- Include the uncomfortable parts — Like "prepare for their defensive reaction" or "know when to escalate to HR"
- Be opinionated — "The exec update is your team's marketing" isn't neutral advice, it's a perspective earned from experience
- Output something usable — Every prompt specifies an output format you can immediately use in a meeting, doc, or conversation
- Acknowledge politics — Real leadership happens in political contexts. These prompts include stakeholder dynamics, not just best practices
Adding Your Own Prompts
Add entries to prompts.json following the existing schema:
CODEBLOCK1
Credits
Created by Rob — engineering manager since 2011, currently running leadingin.tech. These prompts come from real situations managing teams at startups and scale-ups, not from management textbooks.
领导力提示词
经过实战检验的提示词库,专为工程经理、技术负责人、工程副总裁和CTO设计。这些不是泛泛的给我讲讲领导力提示词——而是针对你每周面临的实际挑战,提供具体且有明确观点的框架。
适用人群
- - 管理5-15人团队的工程经理
- 在个人贡献者与管理层边界探索的技术负责人
- 管理经理的总监/副总裁
- 仍亲力亲为的CTO
分类
| 类别 | 数量 | 适用场景 |
|---|
| 一对一会议准备 | 4 | 任何非常规一对一会议前 |
| 团队健康 |
4 | 艰难季度、冲突、裁员或远程脱节后 |
|
事件复盘 | 3 | 事件解决后48小时内 |
|
技术战略 | 4 | 季度规划、架构评审、自研与采购决策 |
|
招聘与面试 | 3 | 开设新岗位、优化流程、说服候选人 |
|
职业发展 | 4 | 晋升周期、反馈传达、留任谈话 |
|
利益相关方沟通 | 4 | 高管汇报、拒绝请求、跨职能对齐、组织重组 |
快速开始
使用命令行
bash
列出所有分类
node scripts/leadership-prompts.js list
获取随机提示词(适合经理技能提升)
node scripts/leadership-prompts.js random
按关键词搜索
node scripts/leadership-prompts.js search 晋升
按ID显示特定提示词
node scripts/leadership-prompts.js show career-dev-promotion
获取某分类下所有提示词
node scripts/leadership-prompts.js category 团队健康
配合AI助手使用
只需告诉你的助手:
我需要准备与一位表现不佳的下属进行一对一谈话。使用领导力提示词技能找到合适的提示词,然后带我一步步完成。
或者浏览时:
显示领导力提示词中所有招聘相关的提示词
如何使用提示词
每个提示词都包含用{花括号}标注的占位变量。填入你的具体情境即可。信息越具体,输出效果越好。
示例:
提示词内容:
我正在准备与一位过去{时间段}表现不佳的下属进行一对一谈话...
你填入:
我正在准备与一位过去6周表现不佳的下属进行一对一谈话...
提示词设计原则
这些提示词的设计目标是:
- 1. 强制结构化——采用框架(SBI、RACI、RAG状态),确保输出可执行而非漫无边际
- 包含棘手部分——如为对方的防御性反应做准备或知道何时升级到HR
- 有明确观点——高管汇报就是你的团队营销不是中性建议,而是来自经验的观点
- 输出可直接使用——每个提示词都指定了输出格式,可在会议、文档或对话中立即使用
- 承认政治因素——真正的领导力发生在政治环境中。这些提示词包含利益相关方动态,而不仅仅是最佳实践
添加自定义提示词
按照现有模式向prompts.json添加条目:
json
{
id: 分类-简短名称,
category: 分类名称,
title: 人类可读的标题,
prompt: 包含{variables}的实际提示词文本,
context: 何时使用此提示词,
output_format: AI应生成的内容格式,
example: 展示实际使用的完整示例
}
致谢
由Rob创建——自2011年起担任工程经理,目前运营leadingin.tech。这些提示词来自在初创公司和成长型企业管理团队的实际经验,而非管理教科书。