Token导航 LogoToken导航

Agent Skill · 研究检索

building-team-culture(建设团队文化)

building-team-culture 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

building-team-culture 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 50,133。

总安装 629周安装 27GitHub Stars 3下载 220核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:building-team-culture(建设团队文化)
来源仓库:https://github.com/oldwinter/skills
仓库路径:skills/building-team-culture
安装命令:
npx skills add https://github.com/oldwinter/skills --skill building-team-culture
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/oldwinter/skills --skill building-team-culture

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息,快速定位候选结果。适合在关键词搜索、任务场景或来源线索不明确时使用。可结合来源仓库和原始 README 核验具体用法。安装前建议确认权限范围、维护状态及是否触发联网或文件读写。

安装命令:npx skills add https://github.com/oldwinter/skills --skill building-team-culture

来源仓库内路径:skills/building-team-culture

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/oldwinter/skills --skill building-team-culture;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

building-team-culture 怎么安装?

在终端执行 npx skills add https://github.com/oldwinter/skills --skill building-team-culture。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

building-team-culture 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

building-team-culture 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

building-team-culture 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

building-team-culture 用的人多吗?

公开数据是累计安装 629 次,近一周安装 27 次,下载 220 次,GitHub 3 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

building-team-culture 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
building-team-culture

建设团队文化 · 本页

6293只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Building Team Culture

Scope

Covers

  • Diagnosing the *current* culture (strengths, gaps, “sacred cows”, and where psychological safety breaks)
  • Articulating culture as an operating system (principles → behaviors → decision rules)
  • Defining team norms (communication, meetings, decisions, feedback, conflict)
  • Designing a lightweight rituals/cadence map that reinforces the culture
  • Planning rollout + reinforcement (coaching model, hiring/onboarding hooks, measurement)

When to use

  • “Create a culture code / values and behaviors for my team.”
  • “Our team norms are unclear—write decision-making + communication norms.”
  • “Psychological safety is low—propose concrete practices and rituals to fix it.”
  • “We’re scaling fast—help us preserve what works and change what doesn’t.”
  • “I’m a new leader—help me listen first and then evolve the culture.”

When NOT to use

  • You need an HR/legal investigation, harassment response, or policy/compliance guidance (involve HR/legal)
  • You need to design a full org restructure, comp bands, or performance management system
  • You need to run user/customer research (use conducting-user-interviews) or design a full survey instrument (use designing-surveys)

Inputs

Minimum required

  • Team context: function, size, seniority mix, reporting line, stage (startup/scale/enterprise)
  • Working model: remote/hybrid/in-office; time zones; any planned org changes
  • Current symptoms with 2–5 examples (e.g., slow decisions, blame, low ownership, stagnation)
  • Desired outcomes: what should be *more true* in 4–12 weeks?
  • Constraints: timeline, leadership support, meeting/time budget, “non-negotiables”
  • Existing artifacts (if any): values, handbook, onboarding, meeting cadences, principles
  • Confidentiality constraints (avoid names/PII; use anonymized examples)

Missing-info strategy

  • Ask up to 5 questions from references/INTAKE.md (3–5 at a time).
  • If specifics are missing, proceed with a default culture OS and clearly label assumptions.
  • Do not request secrets or personally identifying details; ask for redacted summaries instead.

Outputs (deliverables)

Produce a Team Culture Operating System Pack in Markdown (in-chat; or as files if requested):

  1. Culture snapshot (what’s true today; strengths/gaps; root causes; “sacred cows”)
  2. Culture code (v1) (3–7 principles, each with behaviors, “do/don’t”, decision rules, anti-patterns)
  3. Team norms (communication, meetings, decisions, feedback, conflict)
  4. Rituals & cadence map (weekly/monthly/quarterly rituals with purpose + owner)
  5. Rollout + reinforcement plan (socialization, coaching model, hiring/onboarding hooks)
  6. Measurement plan (leading indicators + pulse questions)
  7. Risks / Open questions / Next steps (always)

Templates: references/TEMPLATES.md Expanded guidance: references/WORKFLOW.md

Workflow (7 steps)

1) Intake + constraints + safety

  • Inputs: user context; references/INTAKE.md.
  • Actions: Confirm goals, constraints, and what *must not* change. Identify whether the request includes HR/legal risk; if yes, pause and recommend HR/legal involvement. Clarify what artifacts the user wants (culture code only vs full pack).
  • Outputs: Context snapshot + assumptions/unknowns list.
  • Checks: Decision owner and timeline are explicit; sensitive topics are routed appropriately.

2) “Listen first”: build a culture snapshot (don’t invent culture yet)

  • Inputs: current symptoms; existing artifacts; any examples the user can share.
  • Actions: Summarize what the culture rewards/punishes today. Propose a lightweight “listening tour” plan (questions + who to talk to) if the user hasn’t collected input yet.
  • Outputs: Draft culture snapshot (strengths, gaps, root causes, tensions, sacred cows).
  • Checks: Snapshot is evidence-based (examples), not generic; it distinguishes *stated* vs *lived* culture.

3) Diagnose the few moves that matter

  • Inputs: culture snapshot.
  • Actions: Pick 2–4 priority culture shifts. Identify where “stagnation” exists (lack of visible progress/ambition) and what to change to increase creativity and customer impact. List sacred cows to challenge (and why).
  • Outputs: Prioritized culture focus areas + success signals.
  • Checks: Each focus area has a leading indicator (observable behaviors within weeks).

4) Articulate culture as an operating system (culture code v1)

  • Inputs: focus areas; existing values; constraints.
  • Actions: Write 3–7 principles. For each: definition, behaviors, do/don’t, decision rules, and anti-patterns. Prefer articulating what already works and making gaps explicit.
  • Outputs: Culture code (v1) using references/TEMPLATES.md.
  • Checks: Every principle has behavior-level examples; “culture fit” language is replaced with observable standards.

5) Turn principles into norms + rituals (make it real)

  • Inputs: culture code (v1); team operating reality.
  • Actions: Define explicit norms (communication, meetings, decisions, feedback, conflict). Design rituals that reinforce principles (e.g., weekly customer-impact demo, blameless retro, coaching 1:1s). Assign owners and cadences.
  • Outputs: Team norms + rituals/cadence map.
  • Checks: Rituals have a purpose and an owner; norms reduce ambiguity in common failure modes.

6) Reinforcement plan (coaching > policing)

  • Inputs: culture code + norms + rituals.
  • Actions: Design how the culture will be taught and reinforced: onboarding, hiring signals, promotion expectations, and a lightweight coaching model (peer or craft coaches, not just managers).
  • Outputs: Rollout + reinforcement plan (with a 30/60/90-day view).
  • Checks: Reinforcement mechanisms exist beyond “announce the doc”; responsibilities are assigned.

7) Quality gate + finalize

  • Inputs: full draft pack.
  • Actions: Run references/CHECKLISTS.md and score with references/RUBRIC.md. Add Risks / Open questions / Next steps. Recommend the smallest next experiment (1–2 rituals or norms) to validate impact.
  • Outputs: Final Team Culture Operating System Pack.
  • Checks: Pack is actionable and internally consistent; tradeoffs and risks are explicit.

Quality gate (required)

Examples

Example 1 (new leader, listen-first): “I’m a new Head of Product joining a remote team of 14. Culture feels low-trust and decisions are slow. Create a culture snapshot, a culture code, and explicit decision-making + meeting norms. Include a 30/60/90 rollout and measurement plan.” Expected: full pack with clear norms and rituals; no generic values.

Example 2 (scaling + coaching culture): “We’re growing from 8 → 25. I want to preserve high ownership while adding more coordination. Draft a culture code and a coaching model, plus rituals that keep ambition and creativity high.” Expected: principles + behaviors, coaching model, rituals/cadence map.

Boundary example: “We have a harassment complaint and need to ‘fix our culture’ immediately.” Response: direct to HR/legal for investigation and safety; offer to help later with culture articulation, norms, and reinforcement once appropriate.

平台分布

Codex

38.22%
按下载量换算84

Claude

32.16%
按下载量换算71

Cursor

18.33%
按下载量换算40

Gemini CLI

9.07%
按下载量换算20

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills