Token导航 LogoToken导航

Agent Skill · 研究检索

letter-writer(写信人)

letter-writer 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。

letter-writer 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,权限为权限未标注;许可证为 MIT,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 7,563。

总安装 18,639周安装 503GitHub Stars 公开资料未说明下载 4,740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

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:letter-writer(写信人)
来源仓库:https://github.com/eddiebe147/claude-settings
仓库路径:skills/letter-writer
安装命令:
npx skills add https://github.com/eddiebe147/claude-settings --skill 'Letter Writer'
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/eddiebe147/claude-settings --skill 'Letter Writer'

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。

letter-writer 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。适用于书信模板查询、写作规范参考和文案素材筛选等文字创作支持场景。

通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限范围和联网需求。建议结合原始 README 核验具体用法,注意维护状态及是否触发文件读写操作。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。

来源仓库内路径:skills/letter-writer

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/eddiebe147/claude-settings --skill 'Letter Wri…;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

letter-writer 怎么安装?

在终端执行 npx skills add https://github.com/eddiebe147/claude-settings --skill 'Letter Writer'。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

letter-writer 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。

letter-writer 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。

letter-writer 是免费的吗?

来源标注的许可证是 MIT。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。

letter-writer 用的人多吗?

公开数据是累计安装 18,639 次,近一周安装 503 次,下载 4,740 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。

letter-writer 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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,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 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。
  • 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
letter-writer

写信人 · 本页

18,639—权限未标注MIT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

Letter Writer

The Letter Writer skill automates the creation of professional and personal letters with proper formatting, tone, and structure. It handles various letter types including business correspondence, cover letters, recommendation letters, thank you notes, and formal communications. The skill ensures appropriate formatting, professional language, and customization for different audiences and purposes.

Generate letters in multiple formats (PDF, DOCX, HTML) with customizable templates for different occasions and communication needs.

Core Workflows

Workflow 1: Generate Cover Letter

Purpose: Create a compelling cover letter tailored to a specific job application

Steps:

  1. Collect job details (position, company, requirements)
  2. Extract candidate's relevant qualifications
  3. Structure letter with proper business format
  4. Write engaging opening paragraph
  5. Detail relevant experience and achievements
  6. Connect skills to job requirements
  7. Craft strong closing with call to action
  8. Format with proper spacing and margins
  9. Export to PDF and DOCX

Implementation:

const { Document, Packer, Paragraph, AlignmentType, TextRun } = require('docx');
const fs = require('fs');

async function generateCoverLetter(letterData, outputPath) {
  const doc = new Document({
    sections: [{
      properties: {
        page: {
          margin: {
            top: 1440,    // 1 inch
            right: 1440,
            bottom: 1440,
            left: 1440
          }
        }
      },
      children: [
        // Your contact information
        new Paragraph({
          text: letterData.sender.name,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.sender.address,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: `${letterData.sender.city}, ${letterData.sender.state} ${letterData.sender.zip}`,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.sender.email,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.sender.phone,
          spacing: { after: 200 }
        }),

        // Date
        new Paragraph({
          text: new Date().toLocaleDateString('en-US', {
            year: 'numeric',
            month: 'long',
            day: 'numeric'
          }),
          spacing: { after: 200 }
        }),

        // Recipient information
        new Paragraph({
          text: letterData.recipient.name || 'Hiring Manager',
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.recipient.company,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.recipient.address,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: `${letterData.recipient.city}, ${letterData.recipient.state} ${letterData.recipient.zip}`,
          spacing: { after: 200 }
        }),

        // Salutation
        new Paragraph({
          text: `Dear ${letterData.recipient.name || 'Hiring Manager'}:`,
          spacing: { after: 200 }
        }),

        // Opening paragraph
        new Paragraph({
          text: `I am writing to express my strong interest in the ${letterData.position} position at ${letterData.recipient.company}. ${letterData.opening}`,
          alignment: AlignmentType.JUSTIFIED,
          spacing: { after: 200 }
        }),

        // Body paragraphs
        ...letterData.bodyParagraphs.map(para =>
          new Paragraph({
            text: para,
            alignment: AlignmentType.JUSTIFIED,
            spacing: { after: 200 }
          })
        ),

        // Closing paragraph
        new Paragraph({
          text: letterData.closing || `Thank you for considering my application. I am excited about the opportunity to contribute to ${letterData.recipient.company} and would welcome the chance to discuss how my skills and experience align with your needs. I look forward to hearing from you.`,
          alignment: AlignmentType.JUSTIFIED,
          spacing: { after: 200 }
        }),

        // Signature
        new Paragraph({
          text: 'Sincerely,',
          spacing: { after: 400 }
        }),
        new Paragraph({
          text: letterData.sender.name
        })
      ]
    }]
  });

  const buffer = await Packer.toBuffer(doc);
  fs.writeFileSync(outputPath, buffer);

  return outputPath;
}

// Helper function to generate body paragraphs based on job requirements
function generateCoverLetterBody(candidateData, jobData) {
  const paragraphs = [];

  // Experience paragraph
  const relevantExperience = candidateData.experience
    .filter(exp => exp.relevanceScore > 0.7)
    .slice(0, 2);

  if (relevantExperience.length > 0) {
    const expText = `With ${calculateYearsExperience(candidateData.experience)} years of experience in ${jobData.industry}, I have developed strong expertise in ${jobData.keySkills.slice(0, 3).join(', ')}. In my current role at ${relevantExperience[0].company}, ${relevantExperience[0].achievements[0].toLowerCase()}`;
    paragraphs.push(expText);
  }

  // Skills and qualifications paragraph
  const matchingSkills = candidateData.skills
    .filter(skill => jobData.requiredSkills.includes(skill.name))
    .slice(0, 4);

  if (matchingSkills.length > 0) {
    const skillsText = `I am particularly well-suited for this role given my proficiency in ${matchingSkills.map(s => s.name).join(', ')}. ${candidateData.achievements.find(a => a.includes(matchingSkills[0].name)) || 'I have successfully applied these skills to deliver measurable results.'}`;
    paragraphs.push(skillsText);
  }

  // Company-specific paragraph
  if (jobData.companyInfo) {
    const companyText = `I am particularly drawn to ${jobData.company} because of ${jobData.companyInfo.appeal || 'your reputation for innovation and excellence'}. I believe my background in ${jobData.relevantBackground} would enable me to make immediate contributions to your team.`;
    paragraphs.push(companyText);
  }

  return paragraphs;
}

Workflow 2: Business Letter Generator

Purpose: Create formal business correspondence with proper formatting

Steps:

  1. Select letter type (inquiry, complaint, proposal, etc.)
  2. Format with business letter structure
  3. Use professional tone and language
  4. Include subject line if needed
  5. Write clear, concise content
  6. Add proper closing and signature block
  7. Format for letterhead if available

Implementation:

async function generateBusinessLetter(letterData, outputPath) {
  const letterTypes = {
    inquiry: {
      subject: 'Inquiry Regarding',
      opening: 'I am writing to inquire about',
      tone: 'polite and professional'
    },
    complaint: {
      subject: 'Concern Regarding',
      opening: 'I am writing to bring to your attention',
      tone: 'firm but professional'
    },
    proposal: {
      subject: 'Proposal for',
      opening: 'I am pleased to present',
      tone: 'professional and enthusiastic'
    },
    thankyou: {
      subject: 'Thank You',
      opening: 'I wanted to express my sincere gratitude for',
      tone: 'warm and appreciative'
    }
  };

  const template = letterTypes[letterData.type] || letterTypes.inquiry;

  const doc = new Document({
    sections: [{
      properties: {
        page: {
          margin: { top: 1440, right: 1440, bottom: 1440, left: 1440 }
        }
      },
      children: [
        // Company letterhead (if provided)
        ...(letterData.letterhead ? [
          new Paragraph({
            text: letterData.letterhead.companyName,
            alignment: AlignmentType.CENTER,
            spacing: { after: 0 }
          }),
          new Paragraph({
            text: letterData.letterhead.address,
            alignment: AlignmentType.CENTER,
            spacing: { after: 0 }
          }),
          new Paragraph({
            text: `${letterData.letterhead.phone} | ${letterData.letterhead.email}`,
            alignment: AlignmentType.CENTER,
            spacing: { after: 400 }
          })
        ] : []),

        // Date
        new Paragraph({
          text: new Date().toLocaleDateString('en-US', {
            year: 'numeric',
            month: 'long',
            day: 'numeric'
          }),
          spacing: { after: 200 }
        }),

        // Recipient
        new Paragraph({
          text: letterData.recipient.name,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.recipient.title,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.recipient.company,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.recipient.address,
          spacing: { after: 200 }
        }),

        // Subject line
        ...(letterData.subject ? [
          new Paragraph({
            children: [
              new TextRun({ text: 'RE: ', bold: true }),
              new TextRun(letterData.subject)
            ],
            spacing: { after: 200 }
          })
        ] : []),

        // Salutation
        new Paragraph({
          text: `Dear ${letterData.recipient.name}:`,
          spacing: { after: 200 }
        }),

        // Body
        ...letterData.paragraphs.map(para =>
          new Paragraph({
            text: para,
            alignment: AlignmentType.JUSTIFIED,
            spacing: { after: 200 }
          })
        ),

        // Closing
        new Paragraph({
          text: letterData.closing || 'Sincerely,',
          spacing: { after: 400 }
        }),

        // Signature
        new Paragraph({
          text: letterData.sender.name,
          spacing: { after: 0 }
        }),
        new Paragraph({
          text: letterData.sender.title
        })
      ]
    }]
  });

  const buffer = await Packer.toBuffer(doc);
  fs.writeFileSync(outputPath, buffer);

  return outputPath;
}

Workflow 3: Recommendation Letter Generator

Purpose: Create letters of recommendation for employees, students, or colleagues

Steps:

  1. Collect information about the person being recommended
  2. Understand the purpose and recipient of the letter
  3. Establish credibility and relationship
  4. Detail specific qualifications and achievements
  5. Provide concrete examples
  6. Compare to peers if appropriate
  7. Give enthusiastic endorsement
  8. Offer to provide additional information

Implementation:

function generateRecommendationLetter(recommendationData) {
  const letterData = {
    sender: recommendationData.recommender,
    recipient: recommendationData.recipient || {
      name: 'To Whom It May Concern'
    },
    type: 'recommendation',

    paragraphs: [
      // Opening - establish relationship
      `I am writing to enthusiastically recommend ${recommendationData.candidate.name} for ${recommendationData.purpose}. I have had the pleasure of working with ${recommendationData.candidate.firstName} for ${recommendationData.duration} in my capacity as ${recommendationData.recommender.title} at ${recommendationData.recommender.company}.`,

      // Qualifications and strengths
      `During our time working together, ${recommendationData.candidate.firstName} has consistently demonstrated exceptional ${recommendationData.strengths.join(', ')}. ${recommendationData.standoutQualities}`,

      // Specific examples
      ...recommendationData.examples.map(example =>
        `For instance, ${example.description} This ${example.impact}`
      ),

      // Comparison to peers (if applicable)
      ...(recommendationData.comparison ? [
        `${recommendationData.candidate.firstName} stands out among ${recommendationData.comparison.peerGroup} for ${recommendationData.comparison.distinguishingFactors}.`
      ] : []),

      // Strong endorsement
      `I am confident that ${recommendationData.candidate.firstName} will be an outstanding ${recommendationData.targetRole} and will bring the same level of dedication, skill, and enthusiasm that I have witnessed firsthand. ${recommendationData.candidate.pronoun} has my highest recommendation without reservation.`,

      // Offer additional information
      `Please feel free to contact me at ${recommendationData.recommender.phone} or ${recommendationData.recommender.email} if you would like to discuss ${recommendationData.candidate.firstName}'s qualifications further.`
    ]
  };

  return generateBusinessLetter(letterData, recommendationData.outputPath);
}

Workflow 4: Thank You Note Generator

Purpose: Create professional thank you letters for various occasions

Steps:

  1. Identify occasion (interview, gift, meeting, etc.)
  2. Express gratitude specifically
  3. Mention key details from the interaction
  4. Reinforce interest or relationship
  5. Keep concise but sincere
  6. Use warm but professional tone
  7. Format appropriately for medium

Implementation:

function generateThankYouLetter(thankYouData) {
  const templates = {
    interview: {
      opening: `Thank you for taking the time to meet with me ${thankYouData.when} to discuss the ${thankYouData.position} position at ${thankYouData.company}.`,
      body: `I enjoyed learning more about ${thankYouData.discussionTopics} and was particularly excited to hear about ${thankYouData.highlight}. Our conversation reinforced my strong interest in joining your team and contributing to ${thankYouData.contribution}.`,
      closing: `I am very enthusiastic about this opportunity and believe my ${thankYouData.relevantSkills} would be a strong fit for your needs. Thank you again for your time and consideration. I look forward to hearing from you.`
    },

    meeting: {
      opening: `Thank you for meeting with me on ${thankYouData.date} to discuss ${thankYouData.topic}.`,
      body: `I found our conversation about ${thankYouData.keyPoints} very valuable. Your insights regarding ${thankYouData.insight} gave me a new perspective on ${thankYouData.subject}.`,
      closing: `I appreciate your time and expertise. I look forward to our continued collaboration.`
    },

    gift: {
      opening: `Thank you so much for the thoughtful ${thankYouData.gift}.`,
      body: `Your generosity and thoughtfulness are greatly appreciated. ${thankYouData.personalNote}`,
      closing: `Thank you again for thinking of me. Your kindness means a lot.`
    },

    networking: {
      opening: `It was a pleasure meeting you at ${thankYouData.event} on ${thankYouData.date}.`,
      body: `I enjoyed our conversation about ${thankYouData.topic} and found your perspective on ${thankYouData.insight} particularly interesting. I would love to stay connected and explore opportunities for ${thankYouData.futureCollaboration}.`,
      closing: `Thank you for taking the time to speak with me. I've connected with you on LinkedIn and look forward to staying in touch.`
    }
  };

  const template = templates[thankYouData.type] || templates.meeting;

  return {
    sender: thankYouData.sender,
    recipient: thankYouData.recipient,
    subject: `Thank you - ${thankYouData.regarding}`,
    paragraphs: [
      template.opening,
      template.body,
      template.closing
    ],
    closing: 'Best regards,'
  };
}

Workflow 5: Batch Letter Generation

Purpose: Create multiple personalized letters from a template and data source

Steps:

  1. Load letter template with placeholders
  2. Load recipient data (CSV, JSON, database)
  3. For each recipient:

- Merge data with template - Personalize content - Generate letter - Save with unique filename

  1. Track generation progress
  2. Create summary report

Implementation:

const Handlebars = require('handlebars');

async function batchGenerateLetters(templatePath, recipientsData, outputDir) {
  const template = fs.readFileSync(templatePath, 'utf8');
  const compiledTemplate = Handlebars.compile(template);

  const results = [];

  for (const recipient of recipientsData) {
    try {
      // Merge recipient data with template
      const letterContent = compiledTemplate(recipient);

      // Generate filename
      const filename = `${outputDir}/letter-${recipient.id}-${recipient.name.replace(/\s+/g, '-')}.docx`;

      // Create letter data
      const letterData = {
        sender: recipient.sender || recipientsData.defaultSender,
        recipient: {
          name: recipient.name,
          title: recipient.title,
          company: recipient.company,
          address: recipient.address
        },
        paragraphs: letterContent.split('\n\n').filter(p => p.trim()),
        subject: recipient.subject
      };

      // Generate letter
      await generateBusinessLetter(letterData, filename);

      results.push({
        success: true,
        recipientId: recipient.id,
        name: recipient.name,
        filename: filename
      });

    } catch (error) {
      results.push({
        success: false,
        recipientId: recipient.id,
        name: recipient.name,
        error: error.message
      });
    }
  }

  // Generate summary report
  const summary = {
    total: recipientsData.length,
    successful: results.filter(r => r.success).length,
    failed: results.filter(r => !r.success).length,
    results: results
  };

  fs.writeFileSync(
    `${outputDir}/batch-summary.json`,
    JSON.stringify(summary, null, 2)
  );

  return summary;
}

Quick Reference

ActionCommand/Trigger
Cover letter"write cover letter for [position]"
Business letter"create business letter to [recipient]"
Recommendation"generate recommendation letter"
Thank you note"write thank you letter"
Formal letter"draft formal correspondence"
Batch letters"create letters for [recipients]"

Best Practices

  • Proper Format: Use standard business letter format with correct spacing
  • Professional Tone: Match tone to purpose and audience
  • Concise Writing: Be clear and direct; avoid unnecessary words
  • Proofreading: Zero typos or grammatical errors
  • Personalization: Customize each letter; avoid generic content
  • Specific Details: Include concrete examples and specific information
  • Active Voice: Use active voice for stronger, clearer writing
  • Appropriate Length: Cover letters: 3-4 paragraphs; business letters: as needed but concise
  • Call to Action: End with clear next steps or desired outcome
  • Signature: Include proper signature block with contact information
  • Formatting: Consistent fonts, margins, and spacing
  • Timeliness: Send thank you notes within 24-48 hours

Common Patterns

Job Application Package:

{
  coverLetter: generateCoverLetter(candidateData, jobData),
  resume: generateResume(candidateData),
  references: generateReferencesList(candidateData.references)
}

Customer Service Response:

{
  type: 'response',
  tone: 'apologetic and solution-oriented',
  structure: ['acknowledge issue', 'explain what happened', 'describe resolution', 'offer compensation', 'prevent future']
}

Dependencies

Install required packages:

npm install docx
npm install handlebars     # For templating
npm install pdf-lib        # For PDF generation
npm install nodemailer     # For email sending (optional)

Error Handling

  • Missing Data: Provide default values for optional fields
  • Invalid Addresses: Validate address formatting
  • Template Errors: Catch rendering errors gracefully
  • File Permissions: Handle write errors

Advanced Features

Email Integration:

async function sendLetter(letterPath, recipientEmail, subject) {
  const transporter = nodemailer.createTransporter(config);

  await transporter.sendMail({
    from: senderEmail,
    to: recipientEmail,
    subject: subject,
    html: '<p>Please find attached letter.</p>',
    attachments: [{ path: letterPath }]
  });
}

Signature Integration:

async function addDigitalSignature(letterPath, signatureImagePath) {
  // Add scanned signature image to letter
  // Position at signature block
}

Template Library:

const letterTemplates = {
  coverLetter: './templates/cover-letter.hbs',
  businessInquiry: './templates/business-inquiry.hbs',
  recommendation: './templates/recommendation.hbs',
  thankYou: './templates/thank-you.hbs'
};

平台分布

Claude Code

29.09%
按下载量换算1,379

OpenCode

22.56%
按下载量换算1,069

Gemini CLI

16.65%
按下载量换算789

Antigravity

11.87%
按下载量换算563

Cursor

8.13%
按下载量换算385

windsurf

3.21%
按下载量换算152

安全审计

Gen Agent Trust Hub

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。

继续浏览同类 Skills