Token导航 LogoToken导航TokenDH.com
待分类需要联网github未标认证来源可访问许可证需确认审计通过

planning-under-uncertainty不确定性下的规划

Agent Skill

planning-under-uncertainty 用于记录任务执行中的错误、用户纠正、经验和能力缺口,适合在 Codex、Claude、Cursor、Gemini CLI 中希望让 Agent 持续沉淀问题、修正和最佳实践时使用。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。

总安装

27,456

周安装

1,146

GitHub Stars

734

下载量

8,888
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:planning-under-uncertainty(不确定性下的规划)
来源仓库:https://github.com/refoundai/lenny-skills
仓库路径:skills/planning-under-uncertainty
安装命令:
npx skills add https://github.com/refoundai/lenny-skills --skill planning-under-uncertainty
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/refoundai/lenny-skills --skill planning-under-uncertainty

简介

使用 44 位产品领导者的自适应框架在不确定的环境中进行产品规划。

  • 强调选择性而非预测:在新信息出现时保持灵活调整,而不是遵循僵化的路线图,这在快速发展的 AI/ML 环境中尤其重要
  • 教授决策框架,包括一类/二类决策(可逆与不可逆)、作为指南针而不是 GPS 的数据,以及与学习而不是日历日期相关的明确决策触发器
  • 涵盖技术未知、市场波动、组织变革和 AI/ML 不可预测性的不确定性诊断,并针对每个问题提供量身定制的规划方法
  • 指出常见的陷阱:过度规划、分析瘫痪、忽略领先指标以及仅根据结果而不是学习价值来判断实验

SKILL.md

Planning Under Uncertainty

Help the user navigate product planning when the future is unclear using adaptive planning frameworks from 44 product leaders.

How to Help

When the user asks for help with planning under uncertainty:

  1. Understand the uncertainty type - Ask what's driving the ambiguity: technical unknowns, market volatility, AI/ML unpredictability, or organizational change
  2. Assess planning horizon - Determine if they need short-term execution tactics or long-term strategic flexibility
  3. Match framework to context - Recommend appropriate planning approaches based on their uncertainty profile
  4. Build in adaptation mechanisms - Help them create checkpoints and decision criteria for pivoting

Core Principles

Embrace optionality over prediction

Amjad Masad: "Being agile, not being stuck with roadmaps, being able to just say, oh, we're just going to switch priorities right away, is going to be super important." In rapidly changing environments like AI, maintain flexibility to pivot when new capabilities emerge rather than committing to rigid long-term plans.

Build buffers for chaos

Upasna Gautam: "Any time we're planning we build in buffers for all of that chaos that's happening on a daily basis." In chaotic environments, planning must include explicit time buffers and contingency plans ranging from days to months depending on scope.

Use data as compass, not GPS

Shaun Clowes: "Data is more like a compass than a GPS. If you look at data as a way of giving you the answer, you're always wrong." Use data to validate or invalidate intuition rather than waiting for it to tell you exactly what to do.

Value learning over winning

Ramesh Johari: "Experimentation was never historically in science about winners and losers... Experimentation is always very hypothesis driven. It's about, what are you learning?" A healthy experimentation culture values learning from "failed" risky bets more than safe, incremental "wins."

Develop reproducible testing processes

Nikita Bier: "Develop a reproducible testing process, and that will actually influence the probability of your success more than anything." Success in uncertain markets is driven by the quality and speed of the testing process rather than the initial idea.

Diagnose before acting in crisis

Alex Hardimen: "There's this incredible humility that was needed to really understand and first diagnose what was actually happening on the platform." Managing through a crisis requires "wartime" humility to accurately diagnose problems before attempting solutions.

Create decision triggers, not fixed plans

Eric Ries: "Give yourself a fixed period of time to take some decisive action and see if it feels better." Build checkpoints into plans where you'll reassess based on what you've learned, not just calendar dates.

Distinguish reversible from irreversible decisions

Claire Hughes Johnson: "Type one, type two decisions. Is it high impact? Is it irreversible? Is it not?" Spend more time on one-way doors and move quickly on reversible decisions that can be adjusted later.

Questions to Help Users

  • "What would need to be true for your current plan to work? Which of those assumptions are you least confident about?"
  • "If this takes twice as long as expected, what would you do differently? What if it takes half as long?"
  • "What's the smallest thing you could ship to learn whether your core hypothesis is correct?"
  • "Is this a one-way door or a two-way door decision?"
  • "What signals would tell you to pivot or kill this initiative?"
  • "How much buffer have you built in for unexpected chaos?"

Common Mistakes to Flag

  • Over-planning - Creating detailed long-term roadmaps that create false confidence and resist necessary pivots
  • Analysis paralysis - Waiting for perfect information instead of making decisions with 70% confidence
  • Ignoring leading indicators - Not tracking intermediate signals that could tell you earlier if you're on track
  • Judging experiments by outcomes alone - Not valuing the learning from "failed" experiments that tested important hypotheses
  • Planning theater - Spending excessive time on documents and processes that don't reduce actual uncertainty

Deep Dive

For all 52 insights from 44 guests, see references/guest-insights.md

Related Skills

  • prioritizing-roadmap
  • running-decision-processes
  • scoping-cutting
  • problem-definition

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.71%
按下载量换算3,441

Claude

29.17%
按下载量换算2,593

Cursor

19.29%
按下载量换算1,714

Gemini CLI

10.24%
按下载量换算910

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills