Token导航 LogoToken导航

Agent Skill · 研究检索

narrative-nonfiction(叙事非小说)

narrative-nonfiction 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。

narrative-nonfiction 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 34,439。

总安装 2,228周安装 91GitHub Stars 5下载 713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:narrative-nonfiction(叙事非小说)
来源仓库:https://github.com/rhavekost/author-toolkit
仓库路径:skills/narrative-nonfiction
安装命令:
npx skills add https://github.com/rhavekost/author-toolkit --skill narrative-nonfiction
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/rhavekost/author-toolkit --skill narrative-nonfiction

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。

叙事非小说用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。它支持基于关键词、任务场景或来源线索进行信息筛选与匹配。可通过 npx skills add 命令从指定 GitHub 仓库安装并使用。

安装前建议确认权限范围和维护状态,注意是否会触发联网或文件操作。narrative-nonfiction 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/narrative-nonfiction

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/rhavekost/author-toolkit --skill narrative-non…;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

narrative-nonfiction 怎么安装?

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

narrative-nonfiction 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。

narrative-nonfiction 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。

narrative-nonfiction 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。

narrative-nonfiction 用的人多吗?

公开数据是累计安装 2,228 次,近一周安装 91 次,下载 713 次,GitHub 5 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。

narrative-nonfiction 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
narrative-nonfiction

叙事非小说 · 本页

2,2285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

Narrative Nonfiction Workshop

Workflow for self-help and prescriptive nonfiction using narrative elements and metaphors to guide reader transformation.

Core concept: Prescriptive advice + storytelling. Reader is protagonist on a journey. Book provides map and tools.

When to Use

This skill is for:

  • ✅ Self-help and prescriptive nonfiction books
  • ✅ Memoirs with lessons or transformation arcs
  • ✅ Books using extended metaphors or narrative framing
  • ✅ Practical guides that include storytelling elements
  • ✅ Reader journey design (before state → after state)

When NOT to Use

This skill is NOT for:

  • ❌ Pure fiction (novels, short stories) - use fiction-workshop instead
  • ❌ Academic writing or research papers - different conventions
  • ❌ Straight journalism or reporting - no transformation arc
  • ❌ Technical documentation or how-to guides without story elements
  • ❌ Business books focused purely on data/case studies without reader journey

Stage 1: Foundation Building

Goal: Establish promise, metaphor system, and transformation arc.

Initial Questions

  1. Target reader? (Demographics + psychographics)
  2. Transformation promise?
  3. Central metaphor/framing?
  4. Reader's "before" and "after" states?
  5. Book's unique angle?
  6. How much outlined vs. drafted?

Core Components

Promise: What reader gains. "This book will help you [transformation] by [method]"

Metaphor: Central metaphor, how it maps to advice, where it helps/misleads

Reader's Journey: Entry point (where they start), pain points, resistance, transformation stages, exit point (who they become)

Twist/Reveal (if applicable): What's revealed, setup needed, how to earn payoff

Use assets/book-blueprint-template.md if needed.

Exit condition: Clear grasp of reader, promise, metaphor, arc.


Stage 2: Chapter Development

Goal: Draft or refine chapters balancing advice, story, and exercises.

Chapter structure: Hook (story/question) → Setup (why this matters) → Content (teaching) → Evidence (stories/research) → Application (exercises) → Bridge (to next)

Writing Modes

Switch between these as needed:

ModeInvocationFocus
Voice Editor"Check voice consistency..."Tone, metaphor alignment, author persona
Content Editor"Evaluate the teaching in..."Clarity, completeness, accuracy
Exercise Designer"Design exercises for..."Practical application, appropriate difficulty
Metaphor Consultant"Check metaphor consistency..."Extended metaphor alignment, avoiding confusion
Reveal Engineer"Set up the reveal..."Foreshadowing, misdirection, payoff

See references/ for detailed guidance on each mode.

Creation Workflow

  1. Purpose Check: Key takeaway? Where in arc? What must reader believe before next chapter?
  2. Outline Beats: Hook options, teaching points (2-4), stories/examples, exercises, bridge
  3. Draft: Write chapter. Use "write like you talk" voice.
  4. Layer Metaphor: Present but not forced.
  5. Add Exercises: Use references/exercise-design.md
  6. Polish: Check voice, pacing, reader energy

Editing Workflow

  1. Read as Target Reader: Engaged? Understand? Believe I can do this? Overwhelmed or ready?
  2. Diagnose: Confusion → clarify | Boredom → add story | Resistance → address objections | Overwhelm → simplify
  3. Invoke Mode: Load relevant reference file
  4. Implement: Use str_replace

Stage 3: Arc Integrity Check

Goal: Verify book works as complete transformation journey.

Read full outline/manuscript for:

  1. Promise Delivery: Book delivers promise? Transformation clear and achievable?
  2. Pacing: Change speed appropriate? Integration plateaus? Energy builds?
  3. Metaphor: Maintained throughout? Breaks or contradicts? Still serves at end?
  4. Reveal (if applicable): Twist earned? Seeds planted? Reframe lands emotionally?
  5. Exercise Progression: Build on each other? Difficulty matches stage? Variety?

Common Issues

SymptomCauseFix
"Too preachy"Not enough story/exampleAdd narrative
"Too abstract"Missing concrete adviceAdd specific how-to
"Overwhelming"Too much per chapterNarrow focus, add chapters
"Why should I care?"Missing pain point connectionOpen with reader's struggle
"I can't do this"Missing scaffoldingAdd smaller steps, examples

Self-Check: Is This Working?

Use these checkpoints to verify you're following the workflow correctly.

After Foundation Building:

  • Can you state the book's promise in one sentence?
  • Can you describe the reader's "before" and "after" states clearly?
  • Do you understand the central metaphor and how it maps to the advice?
  • Can you outline the transformation arc stages without looking at notes?

After drafting a chapter:

  • Does the chapter have all six elements: hook, setup, content, evidence, application, bridge?
  • Is the metaphor present but not forced?
  • Is the voice consistent with previous chapters?
  • Would the target reader understand and believe they can apply this?

After designing an exercise:

  • Does the exercise difficulty match where the reader is in their journey?
  • Can the reader complete it with the knowledge they have so far?
  • Is it specific enough to be actionable (not "improve your productivity" but "track your time for three days")?
  • Does it build on previous exercises?

After invoking a mode:

  • Did you explicitly request "Voice Editor" or "Metaphor Consultant" or specific mode?
  • Is the feedback focused on that mode's domain?
  • Did you avoid mixing concerns (voice + content + exercises all at once)?

Before claiming "done":

  • Does the full arc deliver on the promise made in chapter 1?
  • Is the metaphor consistent throughout (or intentionally evolved)?
  • Do exercises progress logically from simple to complex?
  • If there's a reveal/twist, are seeds planted in earlier chapters?

If you answered "no" to any checkpoint, return to that stage before proceeding.


Common Mistakes

MistakeWhy It HappensFix
Skipping Foundation Building"I just want to start writing"Without clarity on promise, metaphor, and arc, chapters drift. Spend 30 minutes on blueprint—saves hours of rewriting.
Forcing the metaphorTrying to make every sentence fit the frameMetaphor should illuminate, not constrain. Use it where it helps understanding, skip where it doesn't. Natural beats forced.
Too much teaching, not enough storyWanting to share all your knowledgeReaders connect through story first. Aim for 40% story/example, 40% teaching, 20% application. Adjust per chapter needs.
Exercises that don't match reader readinessCopying exercise formats from other booksExercise difficulty must match where reader is in arc. Early chapters = simple reflection. Later chapters = bigger challenges.
Losing voice consistencySwitching between academic and conversational tonePick one voice (usually conversational for self-help) and maintain it. Use "Voice Editor" mode to check consistency. See example below.
Ignoring reader resistanceAssuming reader agrees with premiseAddress objections explicitly. "You might be thinking..." shows you understand their skepticism and builds trust.
Reveal without setupPlanning twist ending but not planting seedsIf book has reframe/reveal, every chapter needs subtle foreshadowing. Use "Reveal Engineer" mode to plant and track seeds.

Example: Voice Consistency

Inconsistent voice (Chapter 1 conversational, Chapter 4 academic):

Chapter 1: "You know that feeling of finishing a 'productive' day without accomplishing anything important? That's your brain telling you something's off."

Chapter 4: "Empirical research demonstrates that individuals who engage in time-blocking methodologies exhibit significantly higher levels of task completion and subjective productivity (Chen et al., 2019)."

Consistent voice (both conversational):

Chapter 4: "Here's what the research shows: people who use time-blocking finish more important work. One study tracked 500 knowledge workers for six months and found time-blockers completed 60% more high-priority projects. Your instinct was right."

The difference: Conversational voice maintains "you/your" address, uses plain language, and connects research to reader experience.

Example: Forced vs. Natural Metaphor

Book metaphor: "Think Like an Architect"

Forced: "When you wake up in the morning, draft your day's blueprint. Pour your foundation coffee. Every moment is a chance to design your life's structure."

Natural: "Architects plan for what they need before breaking ground. When you start a project, identify your requirements first—don't just start building and hope it works out."

The difference: Natural metaphor illuminates specific advice. Forced metaphor tries to shoehorn every detail into the frame.

Example: Exercise Design with Scaffolding

Early chapter exercise (reader just learning concept):

EXERCISE: Track Your Time
This week, log what you actually do for three full workdays.
Just notice. Don't judge yourself or try to optimize yet.
Write down: Where did your time go? When were you most focused?

Later chapter exercise (reader has practiced basics):

EXERCISE: Block Your First Deep Work Session
Choose one morning this week (not your busiest day—start manageable).
1. Identify your most important task
2. Block 90 minutes using the protocol from Chapter 3
3. Practice saying it out loud three times before the actual conversation
4. Deliver the boundary
5. Journal: What happened? How did you feel? What would you do differently?

The difference: Early exercises are observation-only with no pressure. Later exercises build on established skills and ask for action.


Quick Reference Commands

NeedCommand
Start new project"Let's build a blueprint for [project]"
Voice check"Check voice consistency in [chapter]"
Content edit"Evaluate the teaching in [chapter]"
Design exercises"Design exercises for [concept]"
Metaphor check"Check metaphor consistency across [chapters]"
Setup the reveal"Help me plant seeds for [reveal] in [chapter]"
Arc review"Review the transformation arc in [outline/draft]"

Files

  • references/transformation-arc.md - Reader journey structure
  • references/metaphor-consistency.md - Extended metaphor management
  • references/exercise-design.md - Practical application design
  • references/reveal-engineering.md - Twist/reframe setup and payoff
  • references/voice-editing.md - Tone and persona consistency
  • assets/book-blueprint-template.md - Book planning document
  • assets/chapter-template.md - Chapter structure template

平台分布

Codex

34.67%
按下载量换算247

Claude

33.93%
按下载量换算242

Cursor

18.02%
按下载量换算128

Gemini CLI

10.6%
按下载量换算76

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