k2f-case-study

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

K2F Case Study — AI-Native Aesthetic Design

K2F 案例研究——AI 原生美学设计

Purpose

目的

Teach agents to blueprint case studies before layout: show how and why work happened—not a resume repeat—with honest scope, evidence-backed results, and hierarchy that hooks skimmers and rewards deep readers.
教会 Agent 在排版前规划案例研究:展示工作如何以及为何发生——而不是重复简历内容——同时如实界定项目范围、以证据支撑结果,并构建既能吸引快速浏览者、又能为深入阅读者提供更多信息的信息层级。

Built on K2F

基于 K2F 构建

This skill produces case studies as
.K2F
packages. Follow the k2f skill for authoring: unpack or init a workspace, edit the semantic tree, run pack/verify, then export. K2F keeps agents on meaning (section roles, narrative beats, visual placeholders)—not pixel coordinates—while the engine compiles layout, typography, and theme into consistent output across PDF, PPTX, DOCX, and other formats.
Before you build, install the core skill:
npx skills add suzheng/k2f --skill k2f
.
此 skill 生成
.K2F
格式的案例研究包。遵循 k2f skill 进行创作:解包或初始化工作区,编辑语义树,运行 pack/verify,然后导出。K2F 让 Agent 专注于内容含义(章节角色、叙事节点、视觉占位符),而不是像素坐标;同时由引擎编译布局、排版和主题,在 PDF、PPTX、DOCX 等格式中生成一致的输出。
构建前,请安装核心 skill:
npx skills add suzheng/k2f --skill k2f
。

When to Use

适用场景

  • Expanding bullets or slide decks into a portfolio or interview-ready project story.
  • Documenting a client or internal win for sales, marketing, or leadership audiences.
  • Tailoring one project for PM, design, engineering, or marketing emphasis without rewriting from scratch.
  • 将简历要点或幻灯片扩展为适合作品集或面试展示的项目故事。
  • 为销售、营销或管理层受众记录客户或内部成果。
  • 针对 PM、设计、工程或营销重点调整同一个项目,无需从头重写。

Workflow

工作流

Step 1. Generate the content_and_design.md

第 1 步:生成 content_and_design.md

Single blueprint for semantic content and aesthetic intent. Use labeled blocks the k2f skill can map to sections and visuals later.
Sub-steps
  1. Story charter — Primary reader (recruiter skim, hiring manager, client prospect), target read depth (3–5 min essential path vs 10–15 min optional depth), and the one outcome that earns attention. List inputs the user provided; mark gaps instead of inventing metrics, quotes, or artifacts.
  2. Role clarity — Your title, team size, timeline, scope, and your contributions vs collective work. Write contribution claims in first person; attribute team-owned areas explicitly.
  3. Information architecture — Default spine (shorten for skim-only audiences):
    • Hero / quick facts: project name, hook metric or outcome, role, timeline, team—scannable metadata strip.
    • Overview: 2–3 sentences; what was delivered and why it mattered.
    • Problem / challenge: business context, user or customer pain, goals, constraints—why this work was worth doing.
    • Process: research and discovery, options explored (breadth), decisions and rationale (depth), iteration from feedback or tests—highlight pivots and insights, not a diary of every meeting.
    • Solution: walkthrough of what shipped; key features or deliverables tied back to stated pains; note where visuals carry the story (before/after, flows, architecture diagram).
    • Results / impact: quantitative outcomes with timeframe and baseline; qualitative signals; business impact; optional “what we’d do differently.”
    • Reflection (optional for portfolio): learnings, skills grown, influence on later work.
    • Visual asset list: each planned image with caption intent, sensitivity redactions, and whether it is essential or nice-to-have.
  4. Audience emphasis — Adjust depth without changing facts: PM/strategy (prioritization, trade-offs, metrics); design/UX (research, iterations, usability); engineering (architecture, constraints, performance); marketing (audience, creative, ROI). Cut sections that do not serve the charter reader.
  5. Design intent — Document tone, hierarchy, density, and palette for the engine:
    • Tone: credible storyteller—specific, modest where evidence is thin; no hype or generic “passion” filler.
    • Hierarchy: lead with the strongest outcome in hero and overview; process subheads surface decisions and insights; results use numbers as visual anchors.
    • Density: short paragraphs and bullets; one idea per block; tables for metric before/after; push long methodology to optional subsections or appendix intent.
    • Visual rhythm: alternate narrative blocks with figure roles; favor journey (sketch → iteration → final) over a single polished screenshot; consistent figure sizing and explanatory captions in the blueprint.
    • Palette: clean portfolio or editorial (neutral body, one accent for metrics and callouts); WCAG AA; restrained use of pull quotes for user or stakeholder voice.
Rules & constraints
  • Show thinking, not just output: every major solution block links to a stated problem or insight.
  • Evidence ladder: separate fact (user data), interpretation (your reading), and projection (expected impact)—never upgrade guesses to results.
  • Process proportion: enough to prove judgment; omit ceremonial steps with no decision or learning.
  • Contribution honesty: no “we” vagueness for interview-facing work; no solo heroics for team-owned delivery.
  • Skim path: a reader who stops after hero + problem + results still understands value.
Common failure modes
  • Resume bullet paste without narrative arc → add problem stakes and decision moments.
  • Solution gallery with no “so what” → tie each artifact to a pain point or metric.
  • Process encyclopedia → collapse to research insight, option trade-off, chosen path.
  • Missing or vague results → quantify where data exists; label
    NEEDS USER INPUT
    where it does not.
  • Undifferentiated team story → sharpen “I led / I built / I validated” boundaries.
  • Only final mocks → plan before/after or iteration figures in the asset list.
  • Generic case study length → match charter depth; cut reflection if the reader is a prospect skimming one win.
Quality checklist
  • Charter names audience, read depth, and hook outcome.
  • Hero facts match overview and results; no contradictory metrics.
  • Problem states goals and constraints; solution explicitly addresses them.
  • Process includes at least one decision with rationale and one iteration or test signal.
  • Results separate measured vs inferred; timeframe and baseline stated.
  • Visual asset list covers essential story beats with caption intent.
  • Design intent specifies hierarchy for hero, results, and figure rhythm.
创建一份统一的蓝图,用于定义语义内容和美学意图。使用带标签的内容块,k2f skill 之后可将其映射到章节和视觉元素。
子步骤
  1. 故事定位 — 主要读者(快速浏览的招聘人员、招聘经理、潜在客户)、目标阅读深度(3–5 分钟的核心路径与 10–15 分钟的可选深度),以及最值得关注的单一成果。列出用户提供的输入;对于信息缺口应明确标记,而不是编造指标、引语或素材。
  2. 角色界定 — 你的职位、团队规模、时间线、工作范围,以及你的贡献与团队集体工作之间的区别。贡献声明使用第一人称;明确标注属于团队负责的部分。
  3. 信息架构 — 默认主线(面向仅做快速浏览的受众时可缩短):
    • Hero / 关键信息:项目名称、吸睛指标或成果、角色、时间线、团队——便于快速扫读的信息栏。
    • 概览:2–3 句话;说明交付了什么,以及为什么重要。
    • 问题 / 挑战:业务背景、用户或客户痛点、目标、约束——为什么这项工作值得投入。
    • 过程:研究与探索、备选方案(广度)、决策及其理由(深度)、基于反馈或测试进行的迭代——重点呈现方向调整和洞察,而不是每次会议的流水账。
    • 方案:逐步讲解最终交付的内容;将关键功能或交付物与已陈述的痛点关联;注明哪些部分需要视觉素材推进叙事(前后对比、流程、架构图)。
    • 结果 / 影响:注明时间范围和基线的量化成果;定性信号;业务影响;可选的“如果重来会采取哪些不同做法”。
    • 反思(作品集可选):学到的经验、提升了哪些技能,以及如何影响后续工作。
    • 视觉素材清单:每张计划图像的图注意图、敏感性信息脱敏要求,以及其属于必需项还是可选项。
  4. 受众侧重 — 在不改变事实的情况下调整内容深度:PM/战略(优先级、取舍、指标);设计/UX(研究、迭代、可用性);工程(架构、约束、性能);营销(受众、创意、ROI)。删除对故事定位所定义读者无用的章节。
  5. 设计意图 — 为引擎记录语气、信息层级、密度和配色方案:
    • 语气:可信的叙事者——表达具体、保持克制;证据不足时不夸大断言;不要使用炒作或泛泛的“激情”填充内容。
    • 层级:在 Hero 和概览中呈现最强成果;过程小标题突出决策和洞察;使用数字作为结果部分的视觉锚点。
    • 密度:采用短段落和项目符号;每个内容块只表达一个观点;使用表格展示指标前后对比;将冗长的方法论放入可选小节或附录规划中。
    • 视觉节奏:让叙事区块与图示角色交替出现;优先展示演进过程(草图 → 迭代 → 最终成果),而不是单张精美截图;在蓝图中保持图示尺寸一致,并配以说明性图注。
    • 配色:采用简洁的作品集或编辑风格(正文使用中性色,指标和标注使用单一强调色);符合 WCAG AA;克制地使用用户或利益相关者声音的引述。
规则与约束
  • 展示思考过程,而不只是最终输出:每个主要方案区块都应关联到已陈述的问题或洞察。
  • 证据阶梯:区分事实(用户提供的数据)、解读(你的判断)和预测(预期影响)——绝不把猜测升级为结果。
  • 过程比例:内容应足以证明判断力;省略没有决策或学习价值的形式化步骤。
  • 贡献真实性:面向面试的作品不得使用含糊的“我们”;团队共同交付的部分也不能包装成个人英雄叙事。
  • 快速浏览路径:即使读者只阅读 Hero、问题和结果,也应理解项目价值。
常见失败模式
  • 直接粘贴简历要点,却没有叙事弧线 → 补充问题的重要性和关键决策时刻。
  • 只有方案展示,却解释不了“所以呢” → 将每项成果与痛点或指标关联。
  • 过程写成百科全书 → 精简为研究洞察、方案取舍和最终选择。
  • 结果缺失或含糊 → 有数据时进行量化;没有数据时标记
    NEEDS USER INPUT
    。
  • 团队故事缺乏区分 → 明确“我主导 / 我构建 / 我验证”的边界。
  • 只有最终设计稿 → 在素材清单中规划前后对比或迭代过程图。
  • 案例研究篇幅千篇一律 → 匹配故事定位所设定的阅读深度;如果潜在客户只是快速浏览某个成果,则删去反思部分。
质量检查清单
  • 故事定位明确受众、阅读深度和吸睛成果。
  • Hero 关键信息与概览和结果一致;不存在相互矛盾的指标。
  • 问题部分说明目标和约束;方案部分明确回应这些问题。
  • 过程至少包含一项有理由支撑的决策,以及一项迭代或测试信号。
  • 结果区分实测值与推断值;注明时间范围和基线。
  • 视觉素材清单覆盖关键叙事节点,并注明图注意图。
  • 设计意图明确 Hero、结果和图示节奏的层级关系。

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 sections and figure roles from the blueprint, apply design intent, verify, and export the audience-ready case study.
遵循 k2f,根据蓝图实例化章节和图示角色,应用设计意图,完成验证并导出面向目标受众的案例研究。