Token导航 LogoToken导航TokenDH.com
AI 工具只读github未标认证来源可访问clear审计通过

story-sense故事感

Agent Skill

story-sense 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

9,126

周安装

392

GitHub Stars

67

下载量

3,199
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

3

许可证

MIT

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:story-sense(故事感)
来源仓库:https://github.com/jwynia/agent-skills
仓库路径:skills/story-sense
安装命令:
npx skills add https://github.com/jwynia/agent-skills --skill story-sense
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/jwynia/agent-skills --skill story-sense

简介

story-sense 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。

  • 适用于项目管理和团队协作场景,帮助跟踪代码变更和协作进度。
  • 能够分析仓库活动、代码提交历史和协作动态,提供项目状态洞察。
  • 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。
  • 可结合来源仓库和原始 README 继续核验具体用法和实现细节。

SKILL.md

Story Sense: Diagnostic Skill

Identify what state a story is in and what it needs to move forward. This is not a linear process but a diagnostic model: Assess → Diagnose → Intervene → Reassess.

When to Use This Skill

Use this skill when:

  • Writer is stuck and doesn't know why
  • Story feels broken but root cause unclear
  • Need to evaluate story problems systematically
  • Someone asks "what's wrong with my story?"

Do NOT use this skill when:

  • Writer wants you to write the story (use story-collaborator)
  • Writer wants coaching questions only (use story-coach)
  • Publishing/marketing questions (use book-marketing)

Core Principle

Story Sense is the ability to know what any story needs, regardless of its current state or intended medium.

There's no such thing as "stuck." There's only:

  • Not yet having diagnosed the problem
  • Not yet applying the right intervention

The Story States

State 0: No Story (Blank Page)

Symptoms: Nothing exists yet Interventions: story-idea-generator, elemental genres

State 1: Concept Without Foundation

Symptoms: Have idea but world/characters/plot feel thin Interventions: cliche-transcendence, systemic-worldbuilding, key-moments

State 2: World Without Life

Symptoms: Setting exists but feels like backdrop Interventions: worldbuilding skill suite (belief-systems, economic-systems, governance-systems)

State 3: Flat Non-Humans

Symptoms: Aliens/fantasy species feel like humans in costume Interventions: conlang, species development frameworks

State 4: Characters Without Dimension

Symptoms: Characters serve plot rather than driving it Interventions: character-arc, underdog-unit, positional-revelation

State 4.5: Plot Without Pacing

Symptoms: Scenes work individually but don't accumulate Interventions: scene-sequencing

State 5: Plot Without Purpose

Symptoms: Events happen but don't accumulate meaning Interventions: moral-parallax, key-moments

State 5.5: Dialogue Feels Flat

Symptoms: Characters sound alike, conversations lifeless Interventions: dialogue

State 5.75: Ending Doesn't Land

Symptoms: Story builds well but resolution disappoints Interventions: endings

State 5.85: Draft Not Progressing

Symptoms: Planning done but draft isn't happening Interventions: drafting

State 5.9: Prose Feels Flat

Symptoms: Story works but sentences are functional not memorable Interventions: prose-style

State 6: Draft Complete, Needs Revision

Symptoms: Draft exists but revision feels overwhelming Interventions: revision

State 7: Ready for Evaluation

Symptoms: Story exists but quality uncertain Interventions: sensitivity-check, story-analysis

Decision Tree

Is there anything on the page?
├── NO → story-idea-generator
└── YES → What's the problem?
    ├── Feels generic → cliche-transcendence
    ├── World feels thin → worldbuilding
    ├── Non-humans feel fake → conlang
    ├── Characters flat → character-arc
    ├── Pacing off → scene-sequencing
    ├── Dialogue wooden → dialogue
    ├── Ending weak → endings
    ├── Meaning unclear → moral-parallax
    ├── Draft not progressing → drafting
    ├── Prose flat → prose-style
    └── Draft needs revision → revision

Diagnostic Process

  1. Listen for symptoms - What are they describing as the problem?
  2. Ask clarifying questions - Get specific about where they're stuck
  3. Identify the state - Match symptoms to state list
  4. Name the diagnosis - Explain what you're seeing
  5. Recommend intervention - Point to specific skill
  6. Offer next steps - What should they try first?

Available Scripts

entropy.ts

Injects creative randomness from curated lists.

deno run --allow-read scripts/entropy.ts lies
deno run --allow-read scripts/entropy.ts disasters --count 3
deno run --allow-read scripts/entropy.ts --combo

Lists: lies, ghosts, disasters, dilemmas, professions, locations, collisions, openings

functions.ts

Generates characters from abstract story functions.

deno run --allow-read scripts/functions.ts
deno run --allow-read scripts/functions.ts --setting scifi
deno run --allow-read scripts/functions.ts healer --setting fantasy

Functions: healer, enforcer, keeper_of_secrets, maker, trader, guide, entertainer, death_worker, transgressor

Anti-Patterns

Prescribing Instead of Diagnosing

Fix: Always ask clarifying questions before diagnosing.

Framework Overload

Fix: Recommend one intervention. Expand after reassessment.

Ignoring Writer's Energy

Fix: Balance diagnostic accuracy with what energizes the writer.

Treating Structure as Story

Fix: Keep asking "Does this feel right?" alongside structural diagnosis.

Related Skills

Routes to all fiction skills based on diagnosed state.

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

Claude Code

32.16%
按下载量换算1,029

OpenCode

26.45%
按下载量换算846

Gemini CLI

17.37%
按下载量换算556

Antigravity

12.22%
按下载量换算391

Cursor

7.48%
按下载量换算239

trae

3.22%
按下载量换算103

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。

来源信息

继续浏览同类 Skills