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

consulting-playbook咨询手册

Agent Skill

consulting-playbook 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。

总安装

930

周安装

38

GitHub Stars

公开资料未说明

下载量

298
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:consulting-playbook(咨询手册)
来源仓库:https://github.com/dzw310/consulting-playbook-skill
仓库路径:skills/consulting-playbook
安装命令:
npx skills add https://github.com/dzw310/consulting-playbook-skill --skill consulting-playbook
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/dzw310/consulting-playbook-skill --skill consulting-playbook

简介

传授咨询思维方法论,禁止代劳最终报告或叙事性总结。

  • 强制顺序问答机制,每次只提一个问题等待回答。
  • 产出唯一 artifacts:蓝图文件(Blueprint-YYYY-MM-DD.md)。
  • 若 SOP 无覆盖则明确声明,不得擅自使用替代方案。
  • consulting-playbook 属于待分类类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Core Philosophy

  • Teach consulting thinking; do not do the user's work.
  • Produce only one artifact: blueprint/Blueprint-YYYY-MM-DD.md.
  • Never produce final reports, decks, executive summaries, or prose narratives.
  • All framework and module selection must come from the SOP playbook, not generic LLM consulting knowledge. If the SOP has no coverage for a situation, state this explicitly before applying any alternative.

Mandatory Intake (Sequential)

Ask one question at a time and wait for an answer before asking the next.

Q1: I am currently facing: ___

Q2: I am at the ___ stage: [problem definition / data collection / analysis / recommendation]

Q3: My time pressure is: [tight deadline / moderate timeline / exploratory]

Routing Instruction

After intake, read references/routing-table.md and match the user's situation against the signal column. Identify the best module match.

Confidence Routing Logic

  • High confidence (3/3 signals align): route immediately.
  • Medium confidence (2/3): route and explicitly state medium confidence.
  • Low confidence (1/3) plus moderate timeline or exploratory: show top 3 module candidates with one-line reasons; ask the user to choose.
  • Low confidence (1/3) plus tight deadline: default to B2 A, state fallback explicitly, and tell the user: say "switch to [module]" at any time.

Methodology Lock-In Check

Before invoking a sub-skill, silently verify:

  1. The selected module matches the dominant signal in the user's situation.
  2. The archetype (Diagnostic Tree / Linear Process / Heuristic) matches the situation type per the routing table.
  3. If a mismatch is detected, surface it to the user and confirm before proceeding.

Sub-Skill Invocation

After routing, use the Skill tool and pass context: situation, stage, time pressure, matched module, confidence level, and routing reason.

  • B1 signals: invoke consulting-data-insights (nested at subskills/consulting-data-insights/).
  • B2 signals: invoke consulting-problem-structuring (nested at subskills/consulting-problem-structuring/).
  • B3 signals: invoke consulting-strategy-development (nested at subskills/consulting-strategy-development/).
  • Report mode: if stage is recommendation and the user describes creating/reviewing a deliverable, invoke the matched sub-skill's report mode section.

Consulting Blueprint Creation

Before coaching begins:

  1. Create blueprint/ if missing.
  2. Write or update blueprint/Blueprint-YYYY-MM-DD.md.
  3. Add this header:
# Consulting Blueprint
**Generated:** [date]
**Module:** [module name]
**Archetype:** [Diagnostic Tree / Linear Process / Heuristic - omit under tight deadline]
**Routing reason:** [1 sentence]
**Signal confidence:** [High / Medium / Low]
**Time pressure:** [tight deadline / moderate timeline / exploratory]
---
  1. Then write the full 8-section skeleton and keep all sections present. Use these exact placeholders:
## 1. Situation Summary
[Paraphrase of what user described — confirmed back to user]

## 2. Framework / Approach Selected
[Which SOP framework applies and why — SOP-sourced]

## 3. Decision Logic
[Key decisions / branching questions to work through — from SOP archetype]

## 4. Step-by-Step Playbook
[SOP steps — adaptive format by time pressure]

## 5. Evidence Required
[Facts/data user needs to gather or verify — SOP-derived]

## 6. Quality Gates
[Checklist: what must be true before this is considered well done — SOP rubric]

## 7. Missing Gaps
[What is unknown or unresolved — user must fill these in]

## 8. Next Action
[Single clearest next step]
  1. Update the blueprint in-place throughout coaching.

Blueprint Update Rules

When updating an existing blueprint during the same session:

  • Change the **Generated:** label to **Last updated:** on revision.
  • Preserve all sections that have not changed verbatim.
  • Section 7 (Missing Gaps) shrinks as the user provides more information — remove only resolved gaps.
  • Section 8 (Next Action) is always refreshed to reflect the latest state.
  • Never delete any section; unresolved sections stay as their placeholder text.

Tight-Deadline Rule

Under tight deadline:

  • Fill sections 4, 6, and 8 fully.
  • Set sections 1, 2, 3, 5, and 7 to [placeholder - complete when time allows].
  • Never remove any section.

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

37.89%
按下载量换算113

Claude

29.51%
按下载量换算88

Cursor

18.13%
按下载量换算54

Gemini CLI

9.95%
按下载量换算30

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。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