Token导航 LogoToken导航

Agent Skill · 待分类

running-offsites(异地运行)

running-offsites 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。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。

running-offsites 是待分类类 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,按安装量在待分类类的 11,209 个 Skill 里排第 4,290。

总安装 774周安装 31GitHub Stars 3下载 250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:running-offsites(异地运行)
来源仓库:https://github.com/oldwinter/skills
仓库路径:skills/running-offsites
安装命令:
npx skills add https://github.com/oldwinter/skills --skill running-offsites
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 running-offsites

这个 Skill 是做什么的

这个 Skill 的分类还没最终确定,用法以来源仓库的说明为准。

用于处理 GitHub 仓库、Issue 和 Pull Request 协作信息。适合围绕代码变更和仓库状态进行整理与分析。可结合来源仓库 README 核验具体用法和权限范围。安装前建议确认是否会触发命令执行或文件读写操作。running-offsites 属于待分类类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/running-offsites

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 running-offsites;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

running-offsites 怎么安装?

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

running-offsites 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。

running-offsites 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。

running-offsites 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。

running-offsites 用的人多吗?

公开数据是累计安装 774 次,近一周安装 31 次,下载 250 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。

running-offsites 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。
  • 需要处理待分类类任务、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。

待分类类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
infsh-cli

infsh CLI · Codex、Claude

4,374,576382权限未标注未明确未收录
ai-video-generation

AI 视频生成 · Codex、Claude

4,313,376382调用外部服务未明确有提醒
lark-openapi-explorer

云雀 OpenAPI 浏览器 · Codex、Claude

2,064,1209,011需要联网未明确有提醒
lark-workflow-standup-report

Lark 工作流程站立报告 · Codex、Claude

2,041,6328,978需要联网未明确通过
use-my-browser

使用我的浏览器 · Codex、Claude

1,848,49140操作浏览器未明确未通过
running-offsites

异地运行 · 本页

7743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

Running Offsites

Scope

Covers

  • Designing an offsite that produces durable alignment + decisions + momentum (not “good vibes”)
  • Choosing the right format: 1-day whiteboard sprint, multi-day retreat, or remote-first quarterly bursts
  • Creating conditions for deep work: laptops down, physical collaboration, and “out of day-to-day” space
  • Building human connection before hard strategy work (so the room can disagree productively)
  • Ending with clear artifacts: decisions, action plan, and follow-up communication that prevents backsliding

When to use

  • “Plan a team offsite / retreat with a concrete agenda and outputs.”
  • “Design a strategy offsite that results in decisions and owners.”
  • “We’re remote-first—create a quarterly in-person ‘burst’ plan.”
  • “Create the offsite prework + facilitation guide + logistics checklist.”
  • “Turn our offsite into a documented plan and follow-up comms.”

When NOT to use

  • You’re trying to fix deep interpersonal conflict with a workshop (“storytelling” won’t fix trust); do conflict repair first (or involve HR/leadership).
  • You don’t know what you’re trying to achieve (use problem definition or product vision first).
  • You need a large-scale event with complex vendor/contracting needs (use professional event planning support; this skill focuses on team-level offsites).
  • The request is HR/legal/compliance-sensitive (escalate to humans and follow company policy).

Inputs

Minimum required

  • Offsite type: strategy / planning / retro / team reset / “burst”
  • Outcomes: 3–7 desired outputs (decisions, plans, relationship goals)
  • Participants (count, roles, time zones) + who must be in the room for decisions
  • Constraints: dates, duration, budget band, location constraints, accessibility needs
  • Current state: what’s broken today + any known tensions/elephants-in-the-room

Missing-info strategy

  • Ask up to 5 questions from references/INTAKE.md (3–5 at a time).
  • If details are missing, proceed with explicit assumptions and offer 2–3 format options.
  • Do not request secrets; use anonymized/redacted context.

Outputs (deliverables)

Produce an Offsite Pack (Markdown in-chat; or files if requested) in this order:

  1. Offsite Brief (1-pager) (purpose, outcomes, decisions-in-scope, attendees, constraints, success measures)
  2. Agenda (timed) + Session Output Map (each block has an explicit artifact output)
  3. Prework Pack (pre-read + prompts + assignments + data/inputs to bring)
  4. Facilitation Run-of-Show (scripts, norms, “laptops down” rules, decision capture method)
  5. Logistics Plan + Checklist (venue/rooms/materials/food/tech/accessibility + backup plan)
  6. Post-Offsite Output Pack (notes, decisions, action plan, follow-up comms, review checkpoints)
  7. Risks / Open questions / Next steps (always included)

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

Workflow (8 steps)

1) Intake + boundary check (don’t design the wrong offsite)

  • Inputs: user context; references/INTAKE.md.
  • Actions: Confirm: offsite type, desired outputs, constraints, and whether there’s underlying conflict/trust debt that must be handled first.
  • Outputs: Context snapshot + assumptions/unknowns list + “offsite is appropriate” decision.
  • Checks: You can name the “elephants” (at a safe level) and decide whether to address them directly, defer them, or escalate.

2) Define outcomes and decisions (artifact-first)

  • Inputs: goals; constraints; stakeholder/participant roles.
  • Actions: Convert goals into 3–7 concrete outputs (decisions, plans, principles, team norms). Define what must be decided in the room vs prepared in advance.
  • Outputs: Offsite Brief sections: outcomes, decisions-in-scope, non-goals, success measures.
  • Checks: Each outcome has an owner and a clear “done” definition.

3) Choose the right format (burst vs retreat vs whiteboard day)

  • Inputs: outcomes; time available; work mode (remote/hybrid/on-site).
  • Actions: Pick an offsite shape:

- Whiteboard day for deep strategic work (“defrag the day”, laptops down). - Retreat (1.5–3 days) for reset + strategy + team bonding. - Burst cadence (quarterly) for remote-first high-velocity creative work.

  • Outputs: Format recommendation + draft schedule skeleton (day blocks) + cadence suggestion.
  • Checks: The format matches the work (creative/strategic/relationship) and the energy budget (no 8 hours of talks).

4) Design the agenda as a sequence of outputs

  • Inputs: outcomes; session library in references/WORKFLOW.md.
  • Actions: Build a timed agenda where every session produces an artifact (decision log entry, principles list, strategy bets, roadmap slice, working agreements). Add breaks and “connection first” time.
  • Outputs: Agenda + Session Output Map.
  • Checks: At least 60% of time is interactive/co-creation; sessions have explicit facilitators and outputs.

5) Build prework that protects the offsite (remove day-to-day)

  • Inputs: topics; data needed; participant availability.
  • Actions: Create a prework pack that gathers inputs asynchronously and reduces meeting time wasted on context. Explicitly set “out of routine” expectations (no email/Slack during blocks, if feasible).
  • Outputs: Prework Pack (pre-read + prompts + assignments).
  • Checks: Participants can arrive ready to decide (not just “get informed”).

6) Write the facilitation run-of-show (make it runnable)

  • Inputs: agenda; team norms; facilitation constraints.
  • Actions: Draft the facilitation script: opening framing, norms, “laptops down” rule, whiteboard usage, breakout instructions, decision capture, and how to handle disagreement safely.
  • Outputs: Facilitation Run-of-Show.
  • Checks: A facilitator could run the day with minimal additional context; decision capture is defined.

7) Logistics + infrastructure (make it frictionless)

  • Inputs: format; location; budget band; accessibility needs.
  • Actions: Create a logistics plan and checklist (venue/rooms, supplies, food, travel, tech, accessibility, contingency plans). For bursts: define a repeatable booking/logistics process.
  • Outputs: Logistics Plan + Checklist.
  • Checks: Known failure modes are covered (no markers, bad room setup, no breaks, travel chaos, AV issues).

8) Close with memory + follow-through (prevent “Monday amnesia”)

  • Inputs: notes/outputs; owners; timelines.
  • Actions: Produce a post-offsite pack: decisions, action plan with owners/dates, follow-up comms, and review checkpoints. Add a “memory artifact” (1-page narrative, principles, or photo of boards + summary).
  • Outputs: Post-Offsite Output Pack + comms draft + review plan.
  • Checks: Every decision has an owner and next step; stakeholders not in the room receive a crisp summary.

Quality gate (required)

Examples

Example 1 (strategy offsite): “Plan a 1-day strategy offsite for a 10-person product team (remote-first). Output the offsite brief, agenda + run-of-show, prework, logistics checklist, and post-offsite decision/action templates.” Expected: Offsite Pack with laptops-down whiteboard blocks + clear decision capture.

Example 2 (quarterly bursts): “Design a quarterly 2-day in-person burst cadence for a distributed team (PST/EST/UK). Include repeatable logistics, prework, and a default agenda that balances connection and deep work.” Expected: Burst program plan + default burst agenda + ops checklist.

Boundary example: “Run a storytelling workshop to fix our leadership team’s distrust.” Response: explain that skills workshops won’t fix trust debt; recommend addressing conflict/relationships first, then revisit an offsite focused on working agreements and decision norms.

平台分布

Codex

35.75%
按下载量换算89

Claude

30.56%
按下载量换算76

Cursor

18.81%
按下载量换算47

Gemini CLI

8.12%
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