做什么
帮助工程管理人员衡量和改善团队交付情况——产生一个通用度量衡为何失败的历史,DORA四键度量框架(部署频率,准备时间,变化故障率,MTTR),DevEx的三个维度(feedback回路,认知负载,流量状态),从工程度量衡到商业结果的翻译层,以及一个测量反标列表来避免
技能库 智客分类:运维与云 developer-productivity
帮助工程管理者衡量并改进团队交付——产生一个通用度量衡为何失败的历史,DORA四键度量框架(部署频率,准备时间,改变故障率,MTTR),DevEx的三个维度(feedback回路,认知负载,流量状态),从工程度量衡到商业结果的翻译层,以及一个用于避免的测量反标的列表. 当用户说"我如何衡量生产率","DORA度量衡","速度","周期时间","开发者体验","DevEx","我如何显示我们的团队表现良好","工程的度量","团队动作慢","工程性能"或"连接工程与业务"时使用. 不要用来管理表现不佳的个人,而是使用业绩审查.
官方网址:skills.sh
先看中文介绍;官方 description 原文单独保留,不改写 SKILL.md。
帮助工程管理人员衡量和改善团队交付情况——产生一个通用度量衡为何失败的历史,DORA四键度量框架(部署频率,准备时间,变化故障率,MTTR),DevEx的三个维度(feedback回路,认知负载,流量状态),从工程度量衡到商业结果的翻译层,以及一个测量反标列表来避免
用户称"我如何衡量生产率","DORA度量衡","高速","周期时间","开发者体验","DevEx","我如何展示我们的团队表现良好","工程的度量","团队速度缓慢","工程性能",或"连接工程与业务". 没有
按 Agent Skills 渐进披露:启动时只加载 name 与 description(约 100 token);任务匹配后才读入整份 SKILL.md 正文;scripts/、references/、assets/ 仅在需要时再读。 本文件正文结构:Developer Productivity、Before Starting、Response Style、How to Use This Skill、Default Response Shape、The Problem With Productivity Metrics。
文件分析:除 SKILL.md 外,正文引用了 references/sources.md,属于带资源的技能包,这些文件按需再读。
Helps engineering managers measure and improve team delivery — produces a history of why common metrics fail, the DORA four-key-metrics framework (deployment frequency, lead time, change failure rate, MTTR), DevEx's three dimensions (feedback loops, cognitive load, flow state), a translation layer from engineering metrics to business outcomes, and a list of measurement anti-patterns to avoid. Use when the user says "how do I measure productivity," "DORA metrics," "velocity," "cycle time," "developer experience," "DevEx," "how do I show our team is performing well," "metrics for engineering," "team is slow," "engineering performance," or "connect engineering to business." Do NOT use for managing an underperforming individual — use performance-reviews instead.
Developer ProductivityBefore StartingResponse StyleHow to Use This SkillDefault Response ShapeThe Problem With Productivity MetricsDORA: The Four Key MetricsDevEx: Three Dimensions of Developer ExperienceQualitative Metrics Are Not SoftTying Engineering Metrics to Business OutcomesWhat Not to DoDive Deeper
来源分类:skills.sh agent-skill
namedeveloper-productivitydescriptionreferences/sources.md以下路径提取自原文;文件是否齐全请以来源仓库中的完整目录为准。
具体调用语法与可用工具以目标 Agent 客户端为准。 查看调用机制说明 ↗
先选择目标 Agent 和安装范围,保留技能包的附属文件,安装后检查客户端能否发现该技能。
该技能引用了附属文件,请从来源获取完整目录;仅复制 SKILL.md 可能缺少依赖。
复制安装指令给支持 Agent Skills 的代理,确认其中的目标目录与客户端匹配。
把 Agent Skill「developer-productivity」安装到我的项目:SKILL.md 原文与官方 description 见 https://zicq.com/zh/skills/skl-b4f51ee8dc931bee-%E5%BC%80%E5%8F%91%E8%80%85%E7%94%9F%E4%BA%A7%E5%8A%9B.html 请存为 .cursor/skills/developer-productivity/SKILL.md 或 .claude/skills/developer-productivity/SKILL.md,frontmatter 的 name 与 description 保持原样,不要改写。 该技能还带 scripts/、references/、assets/ 等文件,请从 https://github.com/manager-dot-dev/manager-skills 取完整目录,不要只建一个 SKILL.md。
需要 Node.js 与 npx。先查看仓库技能列表,确认实际名称。
npx skills add 'https://github.com/manager-dot-dev/manager-skills' --list
npx skills add 'https://github.com/manager-dot-dev/manager-skills' --skill 'developer-productivity'
CLI 会交互选择目标 Agent,默认安装到项目;用户级安装使用 -g。先通过查看命令核对仓库内容,再用 npx skills list 检查已安装技能。
Check for EM context first. If .agents/em-context.md exists, read it.
If .agents/em-context.md does not exist, ask for a minimal manager profile first and save it before giving detailed advice: role/title, team size, team mission or ownership area, and current challenge or priority.
If a specific person is central to the conversation and .agents/reports/[name].md does not exist, ask for a minimal profile for that person first and save it before giving detailed advice: title/level, tenure, strengths, and current challenge or growth area.
.agents/em-context.md or .agents/reports/[name].md automatically. Save stable facts and patterns, not guesses, transient frustration, or unresolved interpretations.Keep the first answer concise and useful. Do not dump the whole framework unless the user asks for depth.
Default to:
When helping with productivity, keep the focus on systems, not individual scoring:
If leadership wants a single productivity number, explain the risk and offer a small dashboard of complementary signals instead.
Measuring developer productivity is one of the hardest problems in engineering management. The history of attempts illustrates why: every metric that gets adopted gets gamed or misinterpreted.
The underlying issue: software development is a knowledge work discipline. Unlike factory output, it can't be measured by counting things without losing what actually matters.
The wrong use of metrics: measuring individuals. Any metric applied to individual developers creates perverse incentives — people optimize for the metric at the expense of the actual work. Don't rank engineers by PR count, commit frequency, or story points.
The right use of metrics: identifying system-level friction. Good metrics answer "where is the team slowing down, and why?" — not "who is performing well?"
The most evidence-backed framework for measuring engineering delivery health. Based on research across thousands of organizations, high performers consistently score well on all four:
| Metric | What it measures | High performer benchmark | |---|---|---| | Deployment frequency | How often you deploy to production | Multiple times per day | | Lead time for changes | Commit to production | Less than 1 hour | | Change failure rate | % of deployments causing incidents | 0–15% | | Mean time to recovery (MTTR) | How quickly you recover from incidents | Less than 1 hour |
These metrics correlate strongly with business outcomes (revenue, customer satisfaction, reliability). They measure the delivery system, not individuals.
How to use them as EM:
The DevEx framework (from DX research) focuses on the developer's lived experience rather than system outputs. It organizes friction into three categories:
Feedback loops — When a developer makes a change, how fast do they know if it worked? This includes CI/CD speed, test run time, code review turnaround, and stakeholder feedback speed. Slow feedback loops break concentration and delay learning.
Cognitive load — How much do developers have to keep in their heads to do their work? Complex processes, unclear ownership, undocumented systems, and context switching all increase cognitive load. High cognitive load slows work and increases errors.
Flow state — Can developers get into deep, uninterrupted focus? Flow state requires: blocks of uninterrupted time, fast tooling, clear goals, and low anxiety. Even good feedback loops and low cognitive load won't produce flow if the environment is fragmented.
How to use it: Run a short team exercise — ask engineers to score each dimension (1–5). The lowest-scoring dimension is your most important focus area. The answers often surface specific, actionable problems (e.g., "our CI takes 45 minutes" or "I never know who owns this service").
A common misconception: quantitative metrics are objective and reliable; surveys and qualitative data are fuzzy and unreliable.
This is wrong. Some of the most important productivity signals can only come from humans:
DORA itself uses surveys for several of its four key metrics — including deployment frequency for organizations that can't measure it automatically. Google's research found that self-reported data is highly reliable when questions are specific and objective.
The practical rule: use quantitative metrics to identify where there's a problem; use qualitative data to understand why. Neither alone gives the full picture.
When leadership asks "how is the engineering team doing?", the answer that lands is the one connected to what they care about.
Common business metrics that engineering directly impacts:
| Business metric | Engineering connection | |---|---| | GRR / NRR (customer retention) | Reliability, quality, user experience | | CAC (cost to acquire customers) | Feature velocity — shipping faster reduces sales cycle | | Time to market | Lead time for changes, deployment frequency | | Support cost | Change failure rate, MTTR |
A practical translation example: "Our change failure rate dropped from 22% to 8% this quarter. That means fewer incidents, less time in firefighting mode, and fewer support escalations — which directly reduces support cost and improves retention."
The EM's job is to build this translation layer. Engineering metrics don't automatically tell the business story — you have to connect the dots explicitly and repeatedly.
If the user asks where a framework came from, wants to read the original article, or wants more context on any topic in this skill — read references/sources.md for the full list of source articles (with links) and books.
team-health — Productivity friction and DevEx signals often surface in team health conversationsroadmap-planning — Delivery metrics inform capacity planning and deadline discussionsmeetings — Flow state is the DevEx dimension most directly affected by meeting culture运维与云
用于容器管理,图像操作,调试的基本道克命令和工作流程.
运维与云
从开放的代理技能生态系统中发现并安装技能. 使用时:(1)用户问"我如何做X",X可能拥有现有技能,(2)用户说"为X找到技能"或"是否为X有技能",(3)用户问"你能否做X",X是专门能力,(4)用户想扩展代理能力,(5)用户想搜索工具,模板,或工作流程,(6)用户提到他们希望…
运维与云
构建,部署,评价,优化,微调,管理微软创始代理,模型,资源最终结束. 用法:铸造、azd ai代理、azd 提供/部署、托管代理脚手架/开发/运行/部署/拖放、即时代理创建、创建代理、更新代理、添加代理工具、援引代理、代理.yaml、评估代理、批量活化、连续活化、持续监测、代理C…
运维与云
执行已有.azure/部署计划.md和基础设施文件的ALREADY-PREPARED应用程序的Azure部署。 当用户向 CREATE 请求新应用程序时, 请不要使用此技能 。 这种技能运行了 azd上行, azd 部署, trarraform 应用, 和 az 部署命令带有内置…