Token导航 LogoToken导航

Agent Skill · 研究检索

impactful-writing(有影响力的写作)

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

impactful-writing 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为权限未标注;许可证为 MIT-0,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 27,939。

总安装 3,305周安装 135GitHub Stars 公开资料未说明下载 1,069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
MIT-0
版本
0.1.0

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:impactful-writing(有影响力的写作)
来源仓库:https://github.com/tenequm/impactful-writing
安装命令:
openclaw skills install impactful-writing
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install impactful-writing

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。

impactful-writing 帮助撰写具有情感共鸣与清晰结构的内容,提升读者行动意愿。适用于 OpenClaw 中文案优化、对外稿件撰写或文档可读性增强。通过 clawhub 安装并使用指定命令调用,支持多语言语气适配。需保持事实准确性,避免过度营销或夸大表述引发信任风险。

建议在正式发布前由人类审核以确保品牌一致性。

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 OpenClaw;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 openclaw skills install impactful-writing;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

    权限没有明确标注,第一次使用时只给一个无关紧要的任务,观察它到底读了哪些文件和网络。

常见问题

6 条

impactful-writing 怎么安装?

在终端执行 openclaw skills install impactful-writing。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

impactful-writing 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。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。

impactful-writing 需要什么权限?

权限级别没有明确标注。第一次使用时建议只给无关紧要的任务,观察它实际读写了哪些内容。

impactful-writing 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

impactful-writing 用的人多吗?

公开数据是累计安装 3,305 次,近一周安装 135 次,下载 1,069 次。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。

impactful-writing 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

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 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 OpenClaw,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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
impactful-writing

有影响力的写作 · 本页

3,305—权限未标注MIT-0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

name
impactful-writing
description
Write clear, emotionally resonant, and well-structured content that readers remember and act upon. Use when writing or editing any text—Twitter posts, articles, documentation, emails, comments, updates—for maximum clarity, engagement, and impact.
metadata
version
0.1.1

Impactful Writing

Overview

Transform any content into clear, memorable, and actionable text using research-backed principles that work across all platforms and contexts. This skill synthesizes 50+ years of readability research, neuroscience of memory, and platform engagement studies into practical techniques.

Core insight: The same psychological principles drive engagement everywhere—clarity reduces cognitive load, specificity creates memory, and structure enables scanning. Master these universal patterns and apply them to any writing context.

When to Use This Skill

  • Writing or editing Twitter/X posts, threads, or social content
  • Creating blog posts, Medium articles, or long-form content
  • Drafting documentation, README files, or technical writing
  • Composing emails, Slack messages, or professional communication
  • Writing GitHub comments, PR descriptions, or code reviews
  • Creating update messages, announcements, or change logs
  • Editing any existing content for clarity and impact

Universal Writing Principles

These evidence-based principles work across all platforms and contexts.

1. Clarity Through Simplicity

Sentence length determines comprehension:

  • 14 words: 90%+ reader comprehension
  • 25 words: Difficulty begins
  • 43 words: <10% comprehension

Target: 15-20 words average, 25 words maximum per sentence.

Word choice matters:

  • Simple words process 76% faster than jargon
  • Active voice processes 15-20% faster than passive
  • Concrete beats abstract (activates sensory brain regions)

2. Structure for Scanning

79% of readers scan rather than read. Design for this reality:

  • Front-load key information (inverted pyramid)
  • Use descriptive headings every 3-4 paragraphs
  • Keep paragraphs to 3-5 sentences maximum
  • Use bullet points for 3+ related items
  • Optimal line length: 50-75 characters

3. Emotional Resonance

Stories trigger oxytocin release, enabling empathy and memory formation:

  • Open with a hook (question, surprising fact, brief story)
  • Use sensory, concrete language
  • Create curiosity gaps (specific questions readers want answered)
  • Close with memorable takeaways (recency effect)

4. Specificity Over Abstraction

Specific details outperform vague statements:

  • "45% increase" beats "significant growth"
  • "in 5 minutes" beats "quickly"
  • "10 ways" beats "several ways"
  • Concrete examples beat abstract explanations

Quick Start Workflow

Writing New Content

  1. Define the core message in one sentence
  2. Open with a hook (question, fact, or story)
  3. Structure with headings for scannability
  4. Use short sentences (15-20 words average)
  5. Close with clear takeaway or call-to-action

Editing Existing Content

  1. Read aloud to identify awkward passages
  2. Cut word count by 10-30% without losing meaning
  3. Convert passive to active voice
  4. Replace jargon with simple words
  5. Add structure (headings, bullets, white space)

Platform-Specific Guidance

Twitter/X Posts

  • Optimal length: 71-100 characters for engagement
  • Hook in first line: Must capture in 3 seconds
  • Use numbers: "10 lessons" outperforms "lessons learned"
  • Thread structure: Each tweet must stand alone AND connect

Example transformation:

Before: "I learned a lot from this experience and want to share some thoughts"
After: "5 hard lessons from shipping 10,000 lines of code in 48 hours:"

Blog Posts / Articles

  • Optimal reading time: 7-10 minutes
  • Headings: Every 300-500 words
  • First paragraph: Must deliver the promise
  • Conclusion: Summarize key points, provide clear next step

Technical Documentation

  • Lead with the goal: What will the reader accomplish?
  • Show, don't tell: Working code examples beat explanations
  • Progressive disclosure: Basic → Advanced
  • Consistent terminology: One term per concept

Professional Communication (Email/Slack)

  • Subject lines: Specific over clever ("Q4 Report Draft" > "Quick update")
  • One topic per message: Increases response rate
  • Front-load action items: Don't bury the ask
  • Keep to half-page maximum: Longer = lower read rate

GitHub Comments / PR Descriptions

  • Start with context: What problem does this solve?
  • Use bullet lists: For changes, decisions, trade-offs
  • Include "why": Reasoning > description
  • Be direct but kind: Critique code, not people

The Revision Checklist

Use this checklist for any content revision:

Clarity Pass:
- [ ] Average sentence length < 20 words
- [ ] No sentence > 30 words
- [ ] Passive voice < 10% of sentences
- [ ] Jargon replaced with simple alternatives

Structure Pass:
- [ ] Opening hook captures attention
- [ ] Key message in first paragraph
- [ ] Headings every 3-4 paragraphs (for longer content)
- [ ] Bullet points for lists of 3+ items
- [ ] Clear call-to-action or takeaway at end

Conciseness Pass:
- [ ] Removed "very," "really," "quite," "just"
- [ ] Replaced multi-word phrases with single words
- [ ] Deleted redundant explanations
- [ ] Cut 10-30% from original word count

Word Reduction Patterns

Common phrases to simplify:

WordyConcise
due to the fact thatbecause
in order toto
at this point in timenow
in the event thatif
with regard toabout
a large number ofmany
in spite of the fact thatalthough
for the purpose ofto

Complex words to simplify:

ComplexSimple
utilizeuse
commencestart
terminateend
demonstrateshow
facilitatehelp
subsequentlater
approximatelyabout
endeavortry

Hook Patterns That Work

Question Hook

Opens with a question the reader wants answered:

"What if everything you knew about productivity was wrong?"

Statistic Hook

Opens with surprising data:

"90% of visitors who read your headline also read your CTA—yet most writers spend 10x more time on body copy."

Story Hook

Opens with a brief narrative:

"At 3 AM, with the deploy failing for the sixth time, I realized the bug wasn't in the code."

Declarative Hook

Opens with a bold statement:

"Most advice about writing is wrong. Here's what actually works."

Contradiction Hook

Challenges an assumption:

"The best writers don't write more. They delete more."

Memory and Impact Principles

Content that sticks follows these patterns:

Serial Position Effect

  • First items: ~70% recall (primacy)
  • Last items: ~60% recall (recency)
  • Middle items: ~40% recall

Implication: Put most important points first and last.

Prediction Errors

Violated expectations create distinctive memories:

Before: "The meeting went exactly as planned."
After: "The meeting started with our CEO apologizing. In 15 years, I'd never seen that."

Sensory Language

Activates multiple brain regions:

Before: "The code was messy."
After: "The code sprawled like tangled Christmas lights—one pull and everything breaks."

Common Anti-Patterns

Over-Explaining

Problem: Explaining what readers already know Fix: Assume intelligence, provide only new information

Buried Lede

Problem: Key point in paragraph 3 Fix: Move conclusion to opening, support with details

Wall of Text

Problem: Dense paragraphs without visual breaks Fix: Add headings, bullets, white space

Passive Avoidance

Problem: "Mistakes were made" (who made them?) Fix: "The team missed the deadline" (clear ownership)

Jargon Cascade

Problem: "We synergized cross-functional paradigms" Fix: "We got the teams to work together"

Proven Content Frameworks

AIDA (Attention → Interest → Desire → Action)

Classic persuasion structure that works for any content with a goal:

Attention: "Most developers waste 3 hours/day on preventable bugs."
Interest: "Static analysis catches 85% of these before they ship."
Desire: "Teams using this approach ship 2x faster with fewer incidents."
Action: "Add this one-line config to your CI pipeline."

PAS (Problem → Agitate → Solution)

Effective for blog posts, landing pages, and persuasive content:

Problem: "Your documentation is outdated the moment you write it."
Agitate: "New devs waste days. Senior devs answer the same questions. Nobody trusts the docs."
Solution: "Generate docs from code comments. Always current, always trusted."

BAB (Before → After → Bridge)

Transformation narrative that creates emotional resonance:

Before: "I spent 6 hours debugging a production issue."
After: "Now I catch these problems before they deploy."
Bridge: "Here's the monitoring setup that changed everything."

1-2-3 Structure

For instructional content—simple, scannable, actionable:

1. The Problem: What's wrong and why it matters
2. The Solution: What to do about it
3. The How: Specific steps to implement

Detailed References

For deeper guidance on specific topics:

Quick Reference: The CLEAR Framework

C - Concise: Cut 10-30% without losing meaning L - Lead with value: Key point in first sentence E - Evidence-based: Specific data beats vague claims A - Active voice: Subject-verb-object structure R - Reader-focused: What do they need to know?

Validation: Content Quality Check

After writing, verify:

  1. Core message test: Can you state it in one sentence?
  2. So what test: After each paragraph, can you answer "so what"?
  3. Grandmother test: Would a non-expert understand the main point?
  4. Action test: Does the reader know what to do next?
  5. Cut test: Can you remove any sentence without losing meaning?

If any test fails, revise that section.

平台分布

OpenClaw

70.03%
按下载量换算749

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

权限未标注

当前来源未能明确判断权限范围,默认进入异常复核队列。

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