Back to Details

k2f-cover-letter

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

K2F Cover Letter — AI-Native Aesthetic Design

K2F 求职信——AI 原生美学设计

Purpose

目的

Teach agents to blueprint cover letters in 
content_and_design.md
before layout: what belongs in the letter versus the resume, how to sequence persuasion (hook → proof → fit → close), and aesthetic intent (tone, hierarchy, letter density)—without inventing employers, referrals, metrics, or company facts the user did not supply.
指导 Agent 在排版前，先在 
content_and_design.md
中规划求职信：哪些内容应放入求职信而非简历、如何安排说服顺序（开场 → 证据 → 契合 → 收尾），以及美学意图（语气、层级、信息密度）——同时不得虚构用户未提供的雇主、内推关系、指标或公司事实。

Built on K2F

基于 K2F 构建

This skill produces cover letters as 
.K2F
packages. Follow the k2f skill for authoring: unpack or init a workspace, edit the semantic tree, pack/verify, and export. K2F keeps agents on meaning (letter blocks, salutation, narrative)—not coordinates—while the engine handles layout and typography.
Before you build, install the core skill: 
npx skills add suzheng/k2f --skill k2f
.
此 skill 会将求职信生成为 
.K2F
包。创作时请遵循 k2f skill：解包或初始化工作区，编辑语义树，打包/验证并导出。K2F 让 Agent 专注于含义（信件区块、称呼、叙事），而非坐标；引擎则处理布局和字体排印。
开始构建前，安装核心 skill：
npx skills add suzheng/k2f --skill k2f
。

When to Use

使用场景

  • Writing or rewriting a cover letter for a specific role and company (with or without a full job description).
  • Referral introductions, internal transfers, speculative applications, or career-change narratives that need framing beyond the resume.
  • Adjusting tone for startup vs enterprise or creative vs formal contexts—not resumes, CVs, thank-you notes, or negotiation emails.
  • 为特定岗位和公司撰写或重写求职信（是否提供完整职位描述均可）。
  • 需要超越简历进行叙事构建的内推介绍、内部转岗、探索性申请或转行经历。
  • 针对初创公司与大型企业、创意行业与正式场景调整语气；不适用于简历、CV、感谢信或薪资谈判邮件。

Workflow

工作流

Step 1. Generate the content_and_design.md

第 1 步：生成 content_and_design.md

Single blueprint for letter content and design intent. Use labeled blocks (letterhead, date/recipient, opening, body paragraphs, closing, sign-off) the k2f skill can map later.
Sub-steps
  1. Letter frame — Role title, company, application type (standard, referral, career change, entry-level, return-to-work), target length (250–400 words, one page, typically 3–4 body paragraphs), and tone band (formal, professional-warm, conversational-confident). Note if a job description and resume are attached.
  2. Requirement map — From the JD (or user brief), list top 3–5 must-haves. Mark strengths to prove (with user-supplied metrics) and gaps to address honestly or omit. Pick one primary match for body paragraph 1 and one secondary theme (extra value, gap reframe, or company fit) for body paragraph 2.
  3. Company research layer — Capture one specific, verifiable hook the user provides or confirms: product, launch, mission line, values, or challenge from the posting. If unknown, flag 
    NEEDS USER INPUT
    —do not invent news or culture claims.
  4. Include / cut — Include: hook, role name, two proof paragraphs (outcomes not on the resume), why-here/why-now, active close. Cut: resume summaries, salary, negativity, hobbies, duplicate CV bullets, generic work ethic, “To whom it may concern” when avoidable.
  5. Opening — Pick one hook type in the blueprint: company insight, referral, JD problem-solver, achievement, or industry angle—not “I am writing to apply” or hollow self-praise.
  6. Body — P1: [their need] + [your experience] + [result] from user data. P2: extra value, honest gap reframe, or deeper company tie-in. One quantified outcome when numbers exist; else 
    metric TBD
    —never invent.
  7. Close & recipient — Specific enthusiasm, discussion ask, brief thanks (not passive “hearing from you”). Name hiring manager if known; else 
    Dear Hiring Manager
    or team salutation. Dates and addresses from user input only.
  8. Design intent — Tone: employer-centric, confident, industry-matched (formal vs startup-direct). Hierarchy: hook and role scannable first; body readable; letterhead quiet. Density: one idea per short paragraph; body dominates the page. Rhythm: parallel proof sentences; one honorific and sign-off style throughout.
Rules & constraints
  • Truth gate: employers, dates, referrals, metrics, and company facts trace to user input or explicit stubs.
  • Non-resume gate: every paragraph must add context, narrative, or motivation not obvious from the CV.
  • Specificity gate: company and role names appear correctly; no placeholder company left in final text.
  • Length gate: over 400 words → cut weakest proof or merge fit into opening, not micro-type.
  • Gap gate: acknowledge gaps with evidence and learning path—never fake credentials.
  • Employer-focus gate: more sentences about their needs and your fit than about what you want from the job.
Common failure modes
  • Generic opening or resume recap → hook + narrative the CV cannot carry.
  • Missing or invented company detail → one verified beat or 
    NEEDS USER INPUT
    .
  • JD keyword stuffing, wrong names, passive close, apologetic career-change tone, or unconfirmed referral → fix against user sources before handoff.
Quality checklist
  • Frame, requirement map (strengths/gaps), and hook type documented.
  • Two body beats tie their needs to user evidence; metric or honest TBD.
  • Verified why-here/why-now; active close; cut list for resume-only content.
  • Design intent covers tone, hierarchy, density, sign-off for the full letter.
这是用于信件内容和设计意图的单一蓝图。使用带标签的区块（信头、日期/收件人、开场、正文段落、结尾、署名），供 k2f skill 后续映射。
子步骤
  1. 信件框架——岗位名称、公司、申请类型（标准申请、内推、转行、初级岗位、重返职场）、目标长度（250–400 个单词、一页，通常为 3–4 个正文段落），以及语气区间（正式、专业且温暖、对话式且自信）。注明是否附有职位描述和简历。
  2. 要求映射——从 JD（或用户简要说明）中列出最重要的 3–5 项必备要求。标出需要证明的优势（附用户提供的指标），以及需要诚实说明或省略的缺口。为正文第 1 段选择一个主要匹配点，为正文第 2 段选择一个次要主题（额外价值、缺口重构或公司契合度）。
  3. 公司调研层——记录用户提供或确认的一个具体、可验证的切入点：产品、发布、使命宣言、价值观，或职位发布中提到的挑战。如未知，则标记 
    NEEDS USER INPUT
    ——不要虚构新闻或文化主张。
  4. 纳入 / 删除——纳入：开场切入、岗位名称、两个证据段落（未在简历中呈现的成果）、为何适合这里/为何是现在、主动式收尾。删除：简历概述、薪资、负面内容、兴趣爱好、与 CV 重复的要点、泛泛的职业操守，以及可避免时的“致相关人士”。
  5. 开场——在蓝图中选择一种开场类型：公司洞察、内推、JD 问题解决视角、成就或行业视角——不要使用“我正在写信申请”或空洞的自我表扬。
  6. 正文——P1：[对方的需求] + [你的经验] + [结果]，均来自用户数据。P2：额外价值、诚实的缺口重构，或与公司更深入的联系。如有数字，使用一个量化成果；否则使用 
    metric TBD
    ——绝不虚构。
  7. 收尾与收件人——具体说明求职热忱、提出进一步交流的请求、简短致谢（不要使用被动的“等您回复”）。如已知招聘负责人姓名，则写明；否则使用“尊敬的招聘经理”或面向招聘团队的称呼。日期和地址仅使用用户提供的信息。
  8. 设计意图——语气：以招聘方为中心、自信且符合行业（正式与初创公司直接风格之间取舍）。层级：先确保开场和岗位易于扫读；正文易读；信头低调。密度：每个短段落只表达一个观点；正文占据页面主体。节奏：使用结构平行的证据句；全文统一使用一种敬称和署名风格。
规则与约束
  • 真实性门槛：雇主、日期、内推关系、指标和公司事实必须可追溯至用户输入或明确标注的占位内容。
  • 非简历门槛：每个段落都必须补充 CV 中不明显的情境、叙事或动机。
  • 具体性门槛：公司名称和岗位名称必须准确；最终文本中不得残留占位公司名称。
  • 长度门槛：超过 400 个单词时，删除最薄弱的证据，或将契合点合并至开场，而不是通过缩小字号处理。
  • 缺口门槛：使用证据和学习路径来回应缺口——绝不伪造资历。
  • 雇主聚焦门槛：关于其需求与你契合度的句子，应多于关于你希望从这份工作中获得什么的句子。
常见失败模式
  • 泛泛的开场或简历复述 → 使用开场切入和 CV 无法承载的叙事。
  • 缺少或虚构公司信息 → 使用一个已验证的切入点或 
    NEEDS USER INPUT
    。
  • 堆砌 JD 关键词、名称错误、被动的收尾、带道歉意味的转行语气，或未经确认的内推 → 交付前依据用户来源进行修正。
质量检查清单
  • 已记录信件框架、要求映射（优势/缺口）和开场类型。
  • 两个正文段落将其需求与用户证据相连接；包含量化指标或诚实的 TBD。
  • 已验证为何是这里/为何是现在；采用主动式收尾；列出应删除的纯简历内容。
  • 设计意图涵盖整封信的语气、层级、密度和署名。

Step 2. Use the content_and_design.md and k2f skill to generate a K2F file and export to an appropriate file format.

第 2 步：使用 content_and_design.md 和 k2f skill 生成 K2F 文件并导出为合适的文件格式

Follow k2f to instantiate letter blocks from the blueprint, apply design intent, verify, and export the application-ready letter (typically PDF or DOCX).
遵循 k2f，根据蓝图实例化信件区块，应用设计意图，进行验证，并导出可直接用于申请的求职信（通常为 PDF 或 DOCX）。