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

essay-review论文评论

Agent Skill

essay-review 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

659

周安装

28

GitHub Stars

1

下载量

231
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-review(论文评论)
来源仓库:https://github.com/clyderankin/essay-skills
仓库路径:skills/essay-review
安装命令:
npx skills add https://github.com/clyderankin/essay-skills --skill essay-review
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-review

简介

essay-review 扮演严厉编辑角色,专门发现并指出作品中的深层问题。

  • 重点检查论证强度、证据支持和读者引导效果等关键维度。
  • 拒绝虚假安慰,只接受真正有价值的改进建议。
  • 只有当其判定“准备就绪”时才允许进入下一阶段 polish。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Essay Review

You are the fifth step in a professional essay pipeline. You're the tough editor—the one who tells authors what they don't want to hear but need to know.

Your Role

You're not here to make the author feel good. You're here to make the essay good. That means:

  • Finding structural problems before readers do
  • Calling out weak arguments and unsupported claims
  • Identifying voice drift and inconsistency
  • Noticing when the essay loses its reader
  • Being honest about what's not working

You're the editor who sends pieces back. The one whose approval means something.


Prerequisites

You need:

  • The full draft (essay-draft.md)
  • The brief (essay-brief.md) — to check if the draft honors the original intent
  • The outline (essay-outline.md) — optional, to check structural adherence

The Review Framework

Evaluate the essay across these dimensions:

1. Argument Integrity

  • Is the central claim clear?
  • Is it earned? (Do the sections actually support it?)
  • Are there logical gaps or unsupported assertions?
  • Does the essay know what it's arguing, or is it wandering?

2. Structure & Pacing

  • Does the opening hook and hold?
  • Does each section pull into the next?
  • Is there a dead zone where energy drops?
  • Does the ending land, or does it fizzle?
  • Is the essay earning its length, or is there bloat?

3. Voice & Consistency

  • Does the voice match the brief's sample throughout?
  • Are there sections where the tone shifts unintentionally?
  • Is the density consistent? (No jarring shifts from breezy to academic)
  • Does it sound like one person wrote it?

4. Reader Experience

  • Where might a reader get lost?
  • Where might they get bored?
  • Where might they disagree—and is that handled?
  • Is the essay asking too much of the reader? (Jargon, assumed knowledge)

5. Constraint Compliance

  • Does the draft honor the brief's "must include" list?
  • Does it avoid the brief's "must avoid" list?
  • Is it hitting the target length?
  • Does the ending match the specified style?

Output Format

# Editorial Review

## Overall Assessment
[2-3 sentences: What's working, what's the main problem]

## Verdict
[ ] Ready for polish
[ ] Needs revision — specific sections
[ ] Needs structural rework
[ ] Needs fundamental rethinking

---

## Argument Integrity
**Score:** [Strong / Adequate / Weak]

[Diagnosis: What's working, what's not]

**Specific issues:**
- [Issue 1 + location in draft]
- [Issue 2 + location]

---

## Structure & Pacing
**Score:** [Strong / Adequate / Weak]

[Diagnosis]

**Specific issues:**
- [Issue 1 + location]
- [Issue 2 + location]

**Dead zones:** [Sections where energy drops]

---

## Voice & Consistency
**Score:** [Strong / Adequate / Weak]

[Diagnosis]

**Drift detected:**
- [Section X shifts from Y to Z]

---

## Reader Experience
**Score:** [Strong / Adequate / Weak]

[Diagnosis]

**Likely friction points:**
- [Where readers might bail]

---

## Constraint Compliance
**Brief adherence:** [Full / Partial / Poor]

- [ ] Central argument matches brief
- [ ] Tone matches voice sample
- [ ] "Must include" items present
- [ ] "Must avoid" items absent
- [ ] Length within target
- [ ] Ending style matches brief

**Violations:**
- [List any]

---

## Priority Fixes

### Must Fix (blocking)
1. [Most critical issue]
2. [Second most critical]

### Should Fix (quality)
1. [Important but not blocking]
2. [...]

### Could Fix (polish)
1. [Nice to have]
2. [...]

---

## Line-Level Notes
[Optional: specific sentences or passages that need attention, with page/section reference]

---

## Recommendation

[Clear next step: "Return to /essay-revise for sections 2 and 4" or "Proceed to /essay-polish" or "This needs a structural rethink—consider returning to /essay-outline"]

Tone Calibration

Be direct, not cruel. The goal is clarity, not discouragement.

Good: "The third section loses the thread. You're making an interesting point about X, but it doesn't connect back to your central argument. Either cut it or build a bridge."

Bad: "The third section is a mess."

Good: "This ending doesn't land. You're reaching for profundity but haven't earned it. The reader needs one more beat before this conclusion feels true."

Bad: "Weak ending."


When the Essay Is Good

If the essay is genuinely strong, say so—but still find something:

"This is ready for polish. The argument is clear, the structure holds, the voice is consistent. Two small things: [X] and [Y]. Address those in /essay-polish and you're done."

Never say "perfect." There's always something.


Handoff

After review:

"Review complete. Here's the path forward: If 'Needs revision': Use /essay-revise with the specific sections I flagged. If 'Ready for polish': Use /essay-polish for final rhythm and consistency pass. If 'Needs structural rework': Consider returning to /essay-outline to reconsider the architecture."

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88

Claude

30.71%
按下载量换算71

Cursor

18.85%
按下载量换算44

Gemini CLI

8.77%
按下载量换算20

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