什么是 KPI 追踪器？

KPI 追踪器是一个全面的解决方案，旨在赋能开发者和企业主直接从其编码环境中监控关键成功指标。作为 Openclaw Skills 生态系统的一部分，它提供了一个结构化框架，将 KPI 划分为收入、运营、营销和客户成功等类别。通过自动计算绩效阈值，该技能可确保团队通过主动报告和数据驱动的见解随时了解业务健康状况。

下载入口:https://github.com/openclaw/skills/tree/main/skills/1kalin/afrexai-kpi-tracker

安装与下载

1. ClawHub CLI

从源直接安装技能的最快方式。

2. 手动安装

将技能文件夹复制到以下位置之一

优先级：工作区 > 本地 > 内置

3. 提示词安装

将此提示词复制到 OpenClaw 即可自动安装。

KPI 追踪器 应用场景

• 使用 Markdown 导出功能自动进行每周或每月的业务绩效复盘。
• 实时标记表现不佳的指标，如高客户流失率或低潜在客户转化率。
• 通过自然语言命令记录绩效数值，简化数据录入。
• 进行深入的趋势分析，以确定业务增速和目标达成概率。

1. 智能体读取 kpi-config.json 文件以建立定义、单位和状态阈值。
2. 用户通过智能体输入数据点，这些数据会自动验证并存储在 kpi-data.json 中。
3. 根据要求，该技能会基于预设目标计算每个指标的当前状态（绿色、黄色或红色）。
4. 智能体使用 Markdown 和表情符号生成可视化状态板，以突出显示需要关注的区域。
5. 对于任何被标记为红色的 KPI，该技能会主动提供根因分析提示以协助排查问题。

KPI 追踪器 配置指南

要开始使用 Openclaw Skills 库中的此条目，请在工作区根目录中创建一个 kpi-config.json 文件：

{
  "kpis": [
    {
      "name": "月度经常性收入",
      "category": "revenue",
      "unit": "$",
      "target": 50000,
      "redBelow": 35000,
      "yellowBelow": 45000
    }
  ]
}

确保智能体具有文件系统访问权限，以便管理同一目录中的 kpi-data.json 存储文件。

KPI 追踪器 数据架构与分类体系

所有数据均以人类可读的 JSON 格式存储在本地，便于进行版本控制和人工审计。

KPI Tracker Skill

Track, analyze, and report on Key Performance Indicators for any business.

What It Does

When activated, this skill helps you:

• Define and categorize KPIs (revenue, ops, marketing, customer success)
• Set targets and thresholds (green/yellow/red)
• Generate weekly/monthly KPI reports in markdown
• Flag KPIs that are off-track with root cause prompts
• Store historical data in a simple JSON file for trend analysis

Usage

Tell your agent: "Track these KPIs" or "Give me a KPI report" or "Which metrics are off track?"

Setup

Create kpi-config.json in your workspace:

{
  "kpis": [
    {
      "name": "Monthly Recurring Revenue",
      "category": "revenue",
      "unit": "$",
      "target": 50000,
      "redBelow": 35000,
      "yellowBelow": 45000
    },
    {
      "name": "Customer Churn Rate",
      "category": "customer",
      "unit": "%",
      "target": 3,
      "redAbove": 7,
      "yellowAbove": 5
    }
  ]
}

Recording Data

Say: "Record MRR at $42,000 for this week"

The agent stores entries in kpi-data.json:

{
  "entries": [
    { "kpi": "Monthly Recurring Revenue", "value": 42000, "date": "2026-02-13", "note": "Post-launch week" }
  ]
}

Reports

Say: "KPI report" and the agent generates a formatted status board:

?? KPI Report — Week of Feb 10, 2026

?? Monthly Recurring Revenue: $48,200 (target: $50,000) — 96.4%
?? Customer Churn Rate: 8.1% (target: 3%) — needs attention
?? Lead Conversion Rate: 11% (target: 15%) — trending up from 9%

?? Action needed on 1 red, 1 yellow KPI

Trend Analysis

Say: "Show MRR trend" — the agent reads historical entries and summarizes direction, velocity, and whether you'll hit target at current pace.

How the Agent Should Behave

1. Read kpi-config.json for KPI definitions
2. Read/write kpi-data.json for historical values
3. When asked for a report: calculate status for each KPI, format with color indicators
4. When a KPI is red: proactively suggest investigation areas
5. When recording: validate the value makes sense (e.g., churn can't be negative)

File Locations

• Config: kpi-config.json (workspace root or custom path)
• Data: kpi-data.json (same directory as config)
• Reports: generated on-demand, optionally saved to reports/kpi-YYYY-MM-DD.md

Pro Tip

Pair this with a cron job to generate weekly KPI reports automatically. For deeper business intelligence and pre-built industry KPI templates, check out AfrexAI Context Packs — drop-in configurations that include KPI frameworks for 10+ industries.