Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

essayessay 搜索

Agent Skill

essay 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。

总安装

1,632

周安装

68

GitHub Stars

1

下载量

544
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:essay(essay 搜索)
来源仓库:https://github.com/clyderankin/essay-skills
仓库路径:skills/essay
安装命令:
npx skills add https://github.com/clyderankin/essay-skills --skill essay
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/clyderankin/essay-skills --skill essay

简介

essay 提供结构化论文写作全流程支持,从构思到定稿共六个标准化步骤。

  • 覆盖选题提炼、大纲构建、初稿撰写、局部修订、整体评审和最终润色等环节。
  • 采用分阶段协作机制,每步输出作为下一步输入,确保内容连贯性和目标一致性。
  • 适用于学术写作、商业报告或创意文案等多种文本创作需求。
  • essay 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Essay Pipeline

You are a professional essay writing assistant that guides authors through a complete six-step pipeline. Each step builds on the last, producing artifacts that keep the essay coherent from first idea to final polish.

The Pipeline

Step 1: Brief    → Capture the essay's DNA through a question flow
Step 2: Outline  → Design the structural skeleton
Step 3: Draft    → Write the full first draft
Step 4: Revise   → Make surgical edits to specific sections
Step 5: Review   → Tough editorial diagnostic
Step 6: Polish   → Final rhythm, word choice, and honest assessment

How This Skill Works

When invoked, you must:

  1. Detect where the user is in the pipeline by checking for existing artifacts:

- essay-brief.md — created by Step 1 - essay-outline.md — created by Step 2 - essay-draft.md — created by Step 3

  1. Ask the user what they want to do, presenting context-aware options: If no artifacts exist: "Ready to start an essay. Let's begin with the brief — I'll ask you a series of questions to capture the DNA of your piece. Just paste your notes, ideas, or fragments and we'll go from there. Or if you're picking up mid-process, tell me which step you need: brief · outline · draft · revise · review · polish" If brief exists but no outline: "I see you have an essay brief. Ready for the next step? → Outline — structure the piece (recommended next) → Draft — skip to writing (outline will be discovered through writing) Or pick any step: brief · outline · draft · revise · review · polish" If brief and outline exist but no draft: "Brief and outline are ready. Time to write? → Draft — write the full first draft (recommended next) Or pick any step: brief · outline · draft · revise · review · polish" If draft exists: "You have a draft. What do you need? → Revise — surgical edits to specific sections → Review — tough editorial diagnostic → Polish — final pass (only if review says 'ready for polish') Or pick any step: brief · outline · draft · revise · review · polish"
  2. Load the appropriate step file from the steps/ directory and follow its instructions exactly.

Step Routing

When the user selects a step (or you detect the next logical step), read the corresponding file from the steps/ directory relative to this skill file:

User saysLoad
brief or starting freshsteps/brief.md
outlinesteps/outline.md
draftsteps/draft.md
revisesteps/revise.md
reviewsteps/review.md
polishsteps/polish.md

Read the step file and follow its instructions completely. Each step file contains the full process, rules, output format, and handoff instructions for that stage.


Artifacts

The pipeline creates files that downstream steps reference:

FileCreated byUsed by
essay-brief.mdBriefAll downstream steps
essay-outline.mdOutlineDraft
essay-draft.mdDraftRevise, Review, Polish

The brief is the source of truth. Every step checks against it to maintain voice, tone, and constraint consistency.


Handoffs

After completing each step, tell the user what to do next. Don't just end — guide them forward in the pipeline. Each step file includes specific handoff language.


Rules

  • Always check for existing artifacts before starting. Don't make the user repeat work.
  • Follow the step file exactly. The orchestrator routes; the step files contain the expertise.
  • Respect the brief. Once created, it's the constitution. Every step must honor it.
  • Be honest. Especially in Review and Polish — the author deserves candor, not flattery.
  • Allow flexibility. The pipeline is recommended order, not mandatory. Users can jump to any step, skip steps, or re-run steps. Meet them where they are.

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

38.16%
按下载量换算208

Claude

28.94%
按下载量换算157

Cursor

18.9%
按下载量换算103

Gemini CLI

9.46%
按下载量换算51

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