Token导航 LogoToken导航

Agent Skill · 研究检索

delegating-work(委派工作)

delegating-work 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。

delegating-work 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 50,488。

总安装 612周安装 26GitHub Stars 3下载 214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:delegating-work(委派工作)
来源仓库:https://github.com/oldwinter/skills
仓库路径:skills/delegating-work
安装命令:
npx skills add https://github.com/oldwinter/skills --skill delegating-work
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 delegating-work

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。

delegating-work 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。通过 npx skills add 命令从指定仓库安装并使用。使用前需确认权限范围、维护状态,避免触发联网或文件读写操作。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/delegating-work

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 delegating-work;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

delegating-work 怎么安装?

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

delegating-work 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。

delegating-work 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。

delegating-work 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。

delegating-work 用的人多吗?

公开数据是累计安装 612 次,近一周安装 26 次,下载 214 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。

delegating-work 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
delegating-work

委派工作 · 本页

6123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

Delegating Work

Scope

Covers

  • Delegating a specific project/problem/decision to a direct report (or cross-functional owner)
  • Transferring context without control (clear outcomes + guardrails, not step-by-step instructions)
  • Setting decision rights, check-in cadence, and “in-the-details” quality reviews without micromanaging

When to use

  • “Help me delegate this project/task to someone on my team.”
  • “I’m holding onto too much / I need to give away my Lego.”
  • “I’m worried I’m micromanaging—how do I stay in the details but empower ownership?”
  • “Draft a delegation brief + decision rights + check-in plan.”

When NOT to use

  • The work is primarily a performance problem (use coaching/feedback workflows instead)
  • You need to decide *what* work to do (prioritization/scope first)
  • You lack a clear outcome, constraints, or success criteria (do a quick problem-definition first)

Inputs

Minimum required

  • The work item to delegate (project/problem/decision) + why now
  • Desired outcome (definition of done) + success metrics or acceptance criteria
  • Constraints/non-negotiables (timeline, budget, quality bar, policies, stakeholders)
  • Delegatee context (role, current load, experience level, growth goals if relevant)

Missing-info strategy

  • Ask up to 5 questions from references/INTAKE.md.
  • If details are unavailable, proceed with explicit assumptions and label unknowns.

Outputs (deliverables)

Produce a Delegation Pack (Markdown in-chat, or files if requested) in this order:

  1. Delegation brief (outcome, context, constraints, stakeholders, timeline)
  2. Decision rights + guardrails (what they can decide, escalation triggers, review points)
  3. Context handoff pack (links, background, “known gotchas”, example outputs)
  4. Execution cadence (check-ins, update format, what “good” looks like)
  5. Review plan (how to be in the details without telling them how to do it)
  6. Debrief plan (learning capture + ownership updates)
  7. Risks / Open questions / Next steps (always included)

Templates: references/TEMPLATES.md

Workflow (8 steps)

1) Frame the delegation decision

  • Inputs: Work item + why now; constraints.
  • Actions: Define the outcome, the “why” (business context), and what must not change (non-negotiables). Decide what “good” looks like.
  • Outputs: Draft Delegation Brief (top section).
  • Checks: The outcome is measurable/testable (someone can say “done” unambiguously).

2) Pick the owner + choose the autonomy level

  • Inputs: Candidate owner(s); their experience and growth goals.
  • Actions: Select the delegatee. Set an explicit autonomy level (e.g., “propose + I approve” vs “you decide, inform me”). Call out which parts are “on assignment” vs “high flexibility.”
  • Outputs: Owner + autonomy statement + boundaries.
  • Checks: Both of you can repeat: “What decisions are yours vs mine?”

3) Transfer context (not instructions)

  • Inputs: Background docs; prior decisions; stakeholders; constraints.
  • Actions: Provide full context so the owner can connect the dots. Share the “why”, the tradeoffs, and known pitfalls. Avoid prescribing the exact path.
  • Outputs: Context Handoff Pack.
  • Checks: Owner can explain the problem, constraints, and success criteria in their own words.

4) Define decision rights + guardrails

  • Inputs: Non-negotiables; risks; stakeholders.
  • Actions: Write decision rights, escalation triggers, and review points. Set “red lines” (quality, policy, safety, customer impact).
  • Outputs: Decision Rights + Guardrails.
  • Checks: Escalation triggers are specific (not “if it feels risky”).

5) Align on plan + milestones (owner-led)

  • Inputs: Draft Delegation Brief + guardrails.
  • Actions: Ask the owner to propose a plan, milestones, and tradeoffs. Act as a thought partner: ask questions instead of giving the answer.
  • Outputs: Milestone plan + first-week plan.
  • Checks: Plan has milestones tied to outcomes, not just activities.

6) Set the execution cadence

  • Inputs: Timeline; team routines; stakeholder needs.
  • Actions: Set check-in frequency, update format, and what you want to see (risks, decisions, asks). Establish how you’ll “refuse to rule” unless a trigger is hit.
  • Outputs: Cadence + Update Template.
  • Checks: Check-ins focus on outcomes/risks/decisions, not task-by-task status.

7) Review in the details without micromanaging

  • Inputs: Work artifacts; review points.
  • Actions: Review output quality via artifacts (docs, specs, results) and criteria. When the team is struggling with the *right* problems, step back. When it’s the wrong problem, intervene with clarity on outcome/guardrails.
  • Outputs: Review notes + decisions (approve/adjust/escalate).
  • Checks: Feedback is framed as “quality bar + constraints” (not “do it my way”).

8) Close the loop (debrief + “give away the Lego”)

  • Inputs: Final deliverable; what happened; learnings.
  • Actions: Debrief what worked/what didn’t, update ownership maps, and explicitly acknowledge the new owner. Capture improvements to templates/guardrails.
  • Outputs: Debrief notes + next delegation candidates.
  • Checks: Ownership is durable (not “it snaps back to you” after delivery).

Quality gate (required)

Examples

Example 1: “Delegate discovery + recommendation for improving onboarding activation to my PM.” Expected: delegation brief, decision rights (what PM decides vs escalates), context pack (data + constraints), weekly cadence, review plan for the final recommendation doc.

Example 2: “Delegate building a lightweight internal tool to an engineer, but I’m worried about security/compliance.” Expected: guardrails + escalation triggers, clear non-negotiables, review points for design + launch readiness, and a debrief plan.

Boundary example: “Delegate ‘make the company strategy better.’” Response: require a narrower outcome (decision to support, artifacts, time box) before producing the Delegation Pack.

平台分布

Claude

32.15%
按下载量换算69

Codex

31.54%
按下载量换算67

Cursor

19.39%
按下载量换算41

Gemini CLI

10.07%
按下载量换算22

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