More screenshots
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
DeepSeek
Muse+22 moreClaude CodeCursorCodex CLIDeepSeekMuse
Dots
Manus
Grok
Grok Bot+18 agentsEcommerce Return Reason Classification Drift
Find inconsistencies in return reason codes across stores marketplaces and reporting periods.
$1
Ecommerce Return Reason Classification Drift
Example session with this skill installed
A sudden increase in damaged item returns follows a marketplace codebook change.
Available evidence: return reason codebooks and marketplace returns export (partial extracts only). Additional sources requested: customer service classification notes and reporting change log. Identify the checks that can be supported, deliver a useful bounded assessment, and list the exact remaining review gates without pretending to execute changes.
- Read your context and instructions
- Compiled the ecommerce return reason
Ecommerce Return Reason Classification Drift — EVIDENCE-LIMITED REVIEW
Scenario received
A sudden increase in damaged item returns follows a marketplace codebook change.
Evidence inventory
- S1: return reason codebooks — partial extract described by the requester; original full source not provided here.
- S2: marketplace returns export — partial extract described by the requester; original full source not provided here.
- S3: customer service classification notes — NOT PROVIDED; required to validate dependent findings.
- S4: reporting change log — NOT PROVIDED; required to validate dependent findings.
Case-specific checks
Normalize historical code revisions. INCOMPLETE until matching source anchors can be examined; document the observable facts, counterexamples and decision gate.
2.
Compare marketplace mappings. INCOMPLETE until matching source anchors can be examined; document the observable facts, counterexamples and decision gate.
3.
Identify ambiguous free text codes. INCOMPLETE until matching source anchors can be examined; document the observable facts, counterexamples and decision gate.
4.
Measure shift attributable to taxonomy changes. INCOMPLETE until matching source anchors can be examined; document the observable facts, counterexamples and decision gate.
Deliverables
Return reason mapping matrix. Create an evidence-linked section with observations, missing data, owner and next validation step.
Unclassified case queue. Create an evidence-linked section with observations, missing data, owner and next validation step.
Trend interpretation memo. Create an evidence-linked section with observations, missing data, owner and next validation step.
Acceptance gate
Trends distinguish classification drift from a proven product defect. Do not label this gate passed until supplied source records actually demonstrate it.
Verdict and next action
INCOMPLETE — evidence required. Request S3 and S4 from the authorized owner, reconcile them with S1 and S2, then repeat the four checks and log remaining conflicts. No production change, approval, communication, or test run is claimed.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing returns across Shopify, Amazon, and Walmart often leads to fragmented data where "Defective" in one system means "Damaged" in another. This skill identifies classification drift and taxonomy inconsistencies that skew retail analytics.
What it does
- Normalize historical revisions by aligning legacy return codes with current reporting standards.
- Map marketplace taxonomy to identify where Amazon, eBay, or Shopify reasons conflict.
- Cleanse free-text inputs by extracting specific return reasons from ambiguous customer service notes.
- Quantify taxonomy impact by measuring how much of a trend shift is due to code changes rather than product quality.
How it works
- Source Registration where you provide codebooks, marketplace exports, and change logs for indexing.
- Evidence Mapping to link specific return instances to the provided classification rules.
- Drift Analysis to flag contradictions between different reporting periods or platforms.
- Report Generation producing a mapping matrix and a trend interpretation memo.
Frameworks & tools
Works with CSV/XLSX exports from Shopify, Amazon Seller Central, and ERP systems. Designed for use within Claude or Cursor environments for data analysis.
Why this beats prompting it yourself
Generic prompts often hallucinate trends or ignore the nuances of changing taxonomy. This skill enforces a strict evidence register, ensuring every classification is tied to a specific source rather than an LLM guess.
Use cases
- Auditing a sudden spike in "Damaged" returns after a marketplace API update.
- Merging return data from a new acquisition into a central BI dashboard.
- Investigating if high return rates are caused by product defects or confusing UI labels.
Known limitations
Does not support direct API connections to marketplaces. Requires human review for domain-sensitive outcomes. Manual exports must be provided as text or structured files.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for E-commerce.
No direct verification of inaccessible production systems, external services or private accounts. Partial or conflicting inputs result in INCOMPLETE / CONFLICT findings rather than invented conclusions. Domain-sensitive decisions require qualified owner review.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 26 days ago
Permissions
Provide only materials you are authorized to share. Use a reviewed environment for proposed changes; keep deployment, production modifications, and third-party communication under explicit human approval.