Connect orders, ads, products, refunds, and fulfillment data into an explainable operations chain, avoiding conclusions based only on revenue, clicks, or platform attribution.
All calculations retain data range, time, currency, timezone, formulas, and gaps; facts, inferences, and recommendations are presented separately.
QUESTION
Write the Problem Specifically
First state the business problem, decision-maker, time range, and actionable actions, e.g., locate the source of profit decline instead of generally asking to analyze all data.
Deliverables
Problem and Decision Contract
Confirmation point
Problem corresponds to specific operational actions
CONTRACT
Define Metric Definitions
Unify order status, net sales, refunds, ad spend, currency, timezone, attribution window, and customer de-duplication rules, and record the system source for each data.
Deliverables
Metric Dictionary and Source Table
Confirmation point
The same metric has only one approved definition
RECONCILE
Reconcile Data First
Check missing, duplicate, delayed, and cross-system differences by date, order, product, and channel; unexplained variances enter the exception list, not directly into operational conclusions.
Deliverables
Reconciliation results and variance reports
Confirmation point
Key amounts traceable to original records
DIAGNOSE
Break down growth and profit
Pinpoint sources of change layer by layer from traffic, conversion, basket size, gross margin, refunds, fulfillment, and repurchase, while flagging sample sizes and possible confounders.
Deliverables
Drivers and evidence chain
Confirmation point
Conclusions do not go beyond what the data can support
ACTION
Form actions and review
For each recommendation, specify the owner, resources, expected impact, risks, validation metrics, and review timing, so analyses enter the task system instead of staying as report files.
Deliverables
Priority action list
Confirmation point
Every action has an owner and validation method
FIXED RELEASE
Analysis framework is public; business data is not sent with the package
FORMWEFT clean-room original document Skill. MIT covers only the original documents in this package; the package contains no client data, scripts, or dependencies. The current fixed hash has been statically reviewed; real data access and output conclusions are still reviewed by the business owner.
Not sure where to start? Copy a learning request.
You can download this reviewed version's resource package. If it includes SKILL.md, provide it and the related documentation to your AI tool; first read and check the materials and usage boundaries, and do not run anything automatically.
Prepare assets: start with fictional or de-identified examples.
Read the boundaries: confirm dependencies, inputs, outputs, and human checkpoints.
Review outputs: keep source and failure records before deciding to pilot.
You can also select the text to copy directly; read it first before executing.
This version has been reviewed.1.0.0
Rewritten independently based on the Feishu directory topic clean-room, without copying or redistributing the original package text. Conclusions apply only to the current fixed hash: the archive structure has been parsed, static rule result is eligible, with 6 files and no scripts, dependencies, or install hooks; online boundaries follow the statements in this resource, and files are output only to user-approved locations. Static review cannot replace runtime environment, input data, or later-version rechecks.
Maintainer
FORMWEFT
License
MIT
Source type
Clean-room original rebuild
Network permission
Offline by default
Local programs
None; you may use company-approved spreadsheet tools to process local copies
Review date
2026/09/04
Review scope
fixedHash
true
static
eligible
files
6
scripts
0
dependencies
0
installHooks
0
network
none; no network capability declared
filesystem
outputs only to a user-approved location
contentReview
completed
independentHumanSignoff
not claimed
SHA-256 · current download package94b504ac22daee291fbd8190446b40b1e33e57d82e06acc325674a1cadb1ce93
Tell us your sales channels, ad platforms, accounting methods, data frequency, and decision roles, and we can continue to build a data contract, reconciliation, attribution, financial reporting, and action tracking workbench.