weekly-ops-report

Generates weekly operational KPI reports with driver decomposition and data-quality auditing.

30.5k|3.5k|Updated Jul 4, 2025
One-click install
npx skills add https://github.com/davila7/claude-code-templates --skill weekly-ops-report
Or copy as Structured Prompt for Agentâ–Ľ
Please help me install this Agent Skill.
Skill: weekly-ops-report
Source: https://github.com/davila7/claude-code-templates/tree/main/cli-tool/components/skills/operations/weekly-ops-report
Command: npx skills add https://github.com/davila7/claude-code-templates --skill weekly-ops-report

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Turning raw operational data into a weekly management report often produces chart dumps that bury the signal. This Skill enforces a contract-first report that answers exactly three questions: what changed, where it is concentrated, and what needs a decision.

Core Features & Use Cases

  • Data-Quality Gate: Audits inputs for duplicated rows, missing calendar days, and impossible values before computing any KPI, with findings listed in a footer.
  • Threshold-Based Findings: Computes KPIs against last week and a trailing 8-week baseline, then only reports movements beyond agreed thresholds, decomposed by driver dimension (region, carrier, category).
  • Fixed Report Structure: Assembles headline KPI cards, a maximum of five tagged findings, 13-week trends, attention lists, and a data-quality footer in the same order every week.
  • Use Case: An operations manager feeds weekly order and shipment data to produce a management report highlighting that service level dropped 2 points, driven mainly by one carrier responsible for 60% of the decline, with a suggested next step.

Quick Start

Use the weekly-ops-report skill to turn this week's operational data into a management report with KPI deltas, top findings, and a data-quality footer.

Frequently Asked Questions about weekly-ops-report

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a weekly operations report from raw data?â–Ľ

Start with a data-quality gate checking for duplicates, missing days, and impossible values. Then compute a small KPI set against last week and a trailing 8-week baseline, filter movements by thresholds, decompose each finding by its main dimension, and assemble a fixed-structure report.

What KPIs should a weekly ops report include?â–Ľ

Keep the core set under about six metrics, typically revenue or volume, service level, and stock cover. Each KPI should show this week, versus last week, and versus the trailing 8-week average so one unusual prior week cannot fake a trend.

Why compare against a trailing 8-week average instead of only last week?â–Ľ

Week-over-week comparison alone manufactures fake trends when the prior week was unusual. A trailing 8-week baseline provides a stable reference so reported movements reflect genuine change rather than noise from a single anomalous week.

How many findings should a weekly management report contain?â–Ľ

Cap findings at five, keeping the largest by impact if more qualify. Only movements beyond agreed thresholds become findings, and each must name its driver dimension and share of the move, otherwise readers learn to ignore the report.

What are common pitfalls in automated weekly reporting?â–Ľ

Common pitfalls include week-over-week-only comparisons, unstated metric definitions, commentary mixed into computed statements, and silent denominator changes such as cancelled orders between weeks. Validate by recomputing one headline KPI directly from raw rows before delivery.