Back to Details

k2f-business-report

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

K2F Business Report — AI-Native Aesthetic Design

K2F 商业报告——AI 原生美学设计

Purpose

目的

Teach agents to blueprint business reports before layout: a decision-oriented story from evidence to action, honest about data gaps, and hierarchy that lets executives grasp performance in minutes—not generic prose or chart dumps without interpretation.
指导 Agent 在排版前规划商业报告：围绕“证据→行动”构建面向决策的故事，坦诚呈现数据缺口，并通过信息层级让高管在几分钟内掌握绩效，而不是堆砌缺乏解读的通用文字或图表。

Built on K2F

基于 K2F 构建

This skill produces business reports 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, metrics, recommendations)—not pixel coordinates—while the engine compiles layout, typography, and theme into consistent output across PDF, DOCX, PPTX, and other formats.
Before you build, install the core skill: 
npx skills add suzheng/k2f --skill k2f
.
此 skill 将商业报告产出为 
.K2F
包。编写时遵循 k2f skill：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 专注于含义（章节角色、指标、建议），而不是像素坐标；引擎则将布局、排版和主题编译为一致输出，覆盖 PDF、DOCX、PPTX 等格式。
构建前，安装核心 skill：
npx skills add suzheng/k2f --skill k2f
。

When to Use

使用场景

  • Weekly, monthly, or quarterly performance reviews from spreadsheets, CRM exports, or pasted metrics.
  • Diagnostic reports that name weak areas, severity, and improvement strategies from user-supplied data.
  • Board- or leadership-ready summaries where trends, risks, and next steps must align with one narrative thread.
  • 基于电子表格、CRM 导出文件或粘贴的指标数据，进行周度、月度或季度绩效回顾。
  • 根据用户提供的数据，指出薄弱领域、严重程度和改进策略的诊断报告。
  • 面向董事会或领导层的摘要，确保趋势、风险和后续步骤围绕同一条叙事主线展开。

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. Report charter — Audience (exec, ops, board), period covered, primary question (“Are we on track?” “Where are we leaking?”), and cadence (weekly pulse vs deep quarterly). List data sources the user actually provided; mark gaps instead of inventing figures.
  2. Clarify scope (when inputs are thin) — Which metrics matter, comparison baseline (prior period, target, budget), and whether the reader needs full detail or highlights-only. Do not expand scope beyond the charter.
  3. Information architecture — Default spine (adapt or shorten for weekly pulses):
    • Cover / title block: report name, period, prepared for, date.
    • Executive summary: headline verdict, 3–5 bullet takeaways, top risk and opportunity—numbers only where user data supports them.
    • Key metrics: small set of KPIs with current value, delta vs prior period and vs target; one line of interpretation each.
    • Performance trends: time-series or segment story; prefer chart roles over long paragraphs when shape matters (growth, decline, volatility).
    • Findings / weak areas: prioritized by severity; each finding = observation + business impact + plausible driver (labeled inference if not in data).
    • Recommendations: 3–6 initiatives tied to findings; per initiative—objective, 3–5 prioritized actions, expected impact (H/M/L), horizon, success metrics.
    • Appendix intent (optional): tables, definitions, methodology—only if the audience needs audit trail.
  4. Analytical discipline — Separate fact (from supplied data), interpretation (your reading), and recommendation (proposed action). Flag anomalies and data quality issues. When severity is assigned, state the rule (e.g., negative growth, missed target, bottom-quartile segment). Never fabricate benchmarks or statistics.
  5. Recommendation shape — Each strategy names a clear program, links to a specific weak area, and ends with measurable success criteria. Cut generic advice (“improve marketing”) unless rewritten as concrete actions with owner-type hints when the user named teams.
  6. Design intent — Document tone, hierarchy, density, palette for the engine:
    • Tone: confident, neutral-analytic; suitable for skimming in a meeting. No hype, no apology essays.
    • Hierarchy: executive summary and KPI strip highest contrast; findings use severity as visual rank (critical before minor); recommendations as scannable cards or numbered blocks.
    • Density: bullets and short paragraphs; one idea per block; tables for metric grids; reserve narrative depth for summary and “so what” under each finding.
    • Palette: professional corporate (deep neutral text, one accent for deltas and severity); WCAG AA; charts limited to a coherent family (trend line, ranked bars, contribution)—no decorative dashboard clutter.
Rules & constraints
  • Decision thread: every section answers “what should leadership do differently?”—cut interesting but non-actionable analysis.
  • Severity before volume: fewer, ranked problems beat long undifferentiated lists.
  • Chart vs prose: use chart intent when the insight is shape or comparison; use prose when causality or nuance needs words.
  • Honesty gate: missing data → 
    NEEDS USER INPUT
    ; do not backfill with plausible numbers.
  • Audience fit: weekly reports emphasize delta and actions; quarterly reports add trend context and initiative portfolios.
Common failure modes
  • Data dump without verdict → add executive summary with explicit headline and ranked takeaways.
  • Recommendations disconnected from findings → map each initiative to a finding ID.
  • Vague weakness (“sales could be better”) → replace with metric, period, and threshold.
  • Equal weight to every KPI → designate 4–7 primary metrics; demote the rest to appendix.
  • Essay trends where a line chart tells the story → switch block to chart role + one interpretive sentence.
  • Fabricated growth rates or benchmarks → remove or label as assumption pending user confirmation.
  • Generic strategy boilerplate → rewrite as named program with actions and metrics.
Quality checklist
  • Charter states audience, period, and primary question.
  • Executive summary readable alone; takeaways match body findings.
  • Each KPI has value, comparison, and one-line “so what.”
  • Findings ordered by severity; impact stated for each.
  • Every recommendation links to a finding and includes success metrics.
  • Facts traceable to user inputs; gaps explicitly marked.
  • Design intent specifies hierarchy for summary, KPIs, findings, and recommendations.
使用一份蓝图统一承载语义内容和美学意图。采用带标签的区块，之后 k2f skill 可将其映射到章节和视觉元素。
子步骤
  1. 报告章程——受众（高管、运营、董事会）、涵盖周期、核心问题（“是否按计划推进？”“问题出在哪里？”）以及报告节奏（每周简报还是季度深度分析）。列出用户实际提供的数据源；应标记缺口，不得虚构数字。
  2. 明确范围（输入不足时）——确定哪些指标重要、比较基线（上期、目标或预算），以及读者需要完整细节还是仅需重点内容。不要超出报告章程的范围。
  3. 信息架构——默认主线（周度简报可调整或精简）：
    • 封面/标题区块：报告名称、周期、呈报对象、日期。
    • 执行摘要：核心判断、3–5 条要点、最大风险和机会——仅在用户数据支持时使用数字。
    • 关键指标：少量 KPI，包括当前值、相对上期和目标的变化幅度；每个指标附一句解读。
    • 绩效趋势：时间序列或细分板块的故事；当趋势形态很重要时，优先采用图表角色，而不是长篇段落（增长、下降、波动）。
    • 发现/薄弱项：按严重程度排序；每项发现包括观察结果、业务影响和可能原因（若原因不在数据中，应标注为推断）。
    • 建议：3–6 项与发现相匹配的举措；每项举措包括目标、3–5 个按优先级排列的行动、预期影响（H/M/L）、实施周期和成功指标。
    • 附录意图（可选）：表格、定义、方法论——仅在受众需要审计依据时提供。
  4. 分析纪律——区分事实（来自所提供的数据）、解读（你的分析）和建议（拟采取的行动）。标记异常和数据质量问题。分配严重程度时，应说明判定规则（例如：负增长、未达目标、处于末四分位的业务板块）。绝不编造基准或统计数据。
  5. 建议结构——每项策略都应指向明确的行动方案，关联具体的薄弱项，并以可衡量的成功标准收尾。除非改写为具体行动，并在用户提到相关团队时附上负责人类型提示，否则不要提供“改善营销”之类泛泛的建议。
  6. 设计意图——为引擎记录语气、层级、信息密度和配色：
    • 语气：自信、客观分析，适合会议中快速浏览。不要炒作，也不要撰写致歉式长文。
    • 层级：执行摘要和 KPI 信息条应具有最高对比度；发现按严重程度进行视觉排序（严重问题优先于轻微问题）；建议采用便于快速浏览的卡片或编号区块。
    • 密度：使用项目符号和短段落；每个区块只表达一个观点；指标矩阵使用表格；叙述深度仅保留在摘要以及每项发现下的“所以呢”解读中。
    • 配色：采用专业企业风格（深色中性文字，使用一种强调色标识变化和严重程度）；符合 WCAG AA；图表限制为一致的系列（趋势线、排名条形图、贡献度），避免装饰性仪表盘杂乱。
规则与约束
  • 决策主线：每个章节都应回答“领导层需要做出哪些不同决策？”——删减有趣但无法转化为行动的分析。
  • 先看严重程度，再看数量：少量且排序明确的问题，优于冗长且缺乏区分的清单。
  • 图表与文字：当洞察体现形态或对比时采用图表意图；当因果关系或细微差异需要文字说明时采用叙述。
  • 诚实性门槛：缺少数据 → 
    NEEDS USER INPUT
    ；不得用看似合理的数字填补。
  • 受众适配：周报强调变化和行动；季度报告增加趋势背景和举措组合。
常见失败模式
  • 只有数据堆砌、没有结论 → 添加带有明确核心判断和排序要点的执行摘要。
  • 建议与发现脱节 → 将每项举措映射到对应的发现 ID。
  • 薄弱项表述模糊（“销售表现还可以更好”）→ 替换为具体指标、周期和阈值。
  • 每个 KPI 权重相同 → 指定 4–7 个主要指标；其余指标降级至附录。
  • 用长篇文字描述趋势，而折线图本可清晰呈现 → 将区块切换为图表角色，并添加一句解读。
  • 虚构增长率或基准 → 删除相关内容，或将其标记为待用户确认的假设。
  • 使用通用战略模板 → 改写为包含明确行动和指标的专项方案。
质量检查清单
  • 章程说明受众、周期和核心问题。
  • 执行摘要可独立阅读；要点与正文发现一致。
  • 每个 KPI 都有数值、比较对象和一句“所以呢”解读。
  • 发现按严重程度排序；每项发现均说明影响。
  • 每项建议都关联一项发现，并包含成功指标。
  • 事实可追溯至用户输入；数据缺口有明确标记。
  • 设计意图规定摘要、KPI、发现和建议的层级。

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 文件并导出为适当的文件格式

Follow k2f to instantiate sections and chart roles from the blueprint, apply design intent, verify, and export the executive-ready deliverable.
遵循 k2f，根据蓝图实例化章节和图表角色，应用设计意图，执行验证，并导出可供高管直接使用的交付物。