← Back to Skills Marketplace
mguozhen

Amazon Backend Report

by mguozhen · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
145
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install amazon-backend-report
Description
Amazon Seller Central backend data report auto-analysis agent. Automatically analyze business reports, inventory reports, advertising reports, and financial...
README (SKILL.md)

Amazon Backend Report Analyzer

Automatically analyze Amazon Seller Central reports — business performance, inventory health, advertising, and financial data — and surface the insights that matter most for your business.

Paste your exported report data. The agent processes it, identifies anomalies, highlights opportunities, and outputs a clear performance summary with prioritized action items.

Commands

report business \x3Cdata>            # analyze business report (traffic + conversion)
report inventory \x3Cdata>           # analyze inventory health report
report advertising \x3Cdata>         # analyze advertising performance report
report financial \x3Cdata>           # analyze financial/payment report
report fba \x3Cdata>                 # analyze FBA performance and fees
report returns \x3Cdata>             # analyze returns and refunds report
report search-terms \x3Cdata>        # analyze search term report from ads
report compare \x3Cperiod1> \x3Cperiod2> # compare two time periods
report digest                     # generate weekly management digest
report save \x3Cperiod>              # save analysis to workspace

What Data to Provide

  • Report data — paste CSV/text data from any Seller Central report
  • Time period — which date range the report covers
  • Historical baseline — prior period data for comparison (if available)
  • Goals — your targets (revenue, ACOS, margin, etc.)
  • ASIN list — specific products to focus on

Report Analysis Framework

Business Report Analysis

Key metrics in Business Report:

Column              What to track               Benchmark
Sessions            Unique visitors             Growing week-over-week
Page Views          Total product page views     Ratio to sessions
Buy Box %           % of time you own Buy Box   >90%
Units Ordered       Units sold per period       Week-over-week trend
Unit Session %      Conversion rate             >10% (varies by category)
Ordered Product Sales Revenue by ASIN           Growing trend

Conversion rate diagnostic:

CVR \x3C 5%:    Listing issue — images, bullets, or price problem
CVR 5-10%:   Below average — improve weakest listing element
CVR 10-15%:  Good — optimize further for incremental gains
CVR > 15%:   Excellent — scale traffic with ads

Sessions dropping:

  • External algorithm change
  • Out-of-stock event caused ranking loss
  • Competitor improved listing quality
  • Seasonal demand shift
  • Ad spend reduced

Sessions growing, revenue not growing:

  • Price decreased (more traffic, same revenue)
  • Conversion rate dropped (traffic quality changed)
  • Returns increased (inflating gross revenue, but refunds eating margin)

Inventory Health Report Analysis

Key inventory metrics:

Metric                  Target              Action if violated
Days of Supply          30-60 days          Reorder if \x3C30 days
Inventory Age (0-90d)   >90% of stock       Monitor aging inventory
Inventory Age (91-180d) \x3C10% of stock       Run promotions to clear
Inventory Age (180d+)   \x3C2% of stock        Deep discount or removal
Stranded Inventory      0 units             Fix listing issues immediately
Unfulfillable Units     0 units             Remove or dispose
Excess Inventory        \x3C30 days excess     Reduce reorder quantity

Long-term storage fee risk:

  • Units stored >181 days incur additional fees ($1.50/cubic foot/month)
  • Units stored >365 days: $6.90/cubic foot/month
  • Action: identify aging SKUs and run promotions before 181-day mark

Inventory planning dashboard:

ASIN | Units | Daily Sales | Days Cover | Reorder Point | Order Now?
BXXX |  450  |    15/day   |   30 days  |  150 units    |  ⚠ Yes
BYYY |  800  |    10/day   |   80 days  |  100 units    |  ✓ No
BZZZ |   50  |     5/day   |   10 days  |   75 units    |  🚨 Urgent

Advertising Report Analysis

Search Term Report (most important ad report):

Analysis steps:
1. Sort by Spend descending — find where budget is going
2. Find terms with high spend, zero conversions → Negative match
3. Find terms with conversions, good ACOS → Add to Exact campaign
4. Find terms with high impressions, 0 clicks → Review bid or relevance
5. Compute true ACOS per term: Ad Spend / Ad Sales × 100

ACOS benchmarks by action:

ACOS \x3C 15%: Raise bids — underinvesting in profitable keywords
ACOS 15-25%: Maintain — at break-even or profitable range
ACOS 25-35%: Review — marginal, may be justified for ranking
ACOS > 35%: Reduce bids or pause — unprofitable

Campaign performance buckets:

Bucket                  ACOS    Impressions  Conversions
Stars:                  \x3C20%    High         High        → Scale budget
Cash Cows:              \x3C20%    Low          Moderate    → Increase bids
Potential:              20-35%  High         Low         → Improve listing
Underperformers:        >35%    Any          Low         → Reduce or pause

Financial Report Analysis

Key financial metrics:

Metric                  Formula                           Target
Gross Revenue           Sum of product sales              —
Amazon Fees             Referral + FBA + storage + ads    \x3C40% of revenue
Returns/Refunds         Refund amount / gross revenue     \x3C5%
Net Revenue             Gross - Returns - Fees            —
Net Margin              Net Revenue / Gross Revenue       >20%

Fee percentage benchmarks by category:

Category        Referral    FBA        Total Fees  Net Margin Target
Electronics     8%          $5-8       18-22%      >25%
Home & Kitchen  15%         $4-6       22-26%      >20%
Clothing        17%         $3-5       22-26%      >20%
Beauty          15%         $3-5       21-25%      >22%

Cash flow planning:

Disbursement cycle: Every 14 days (typically)
Reserved amount: Amazon holds ~7 days of sales as reserve
Working capital needed: 45-60 days of COGS (production + freight + reserve)

Weekly Management Digest Format

WEEK OF [DATE] — PERFORMANCE DIGEST

📊 REVENUE
  This week: $XX,XXX | Last week: $XX,XXX | Change: +/-X%
  YTD: $XXX,XXX | YTD goal: $XXX,XXX | On track: Yes/No

📦 INVENTORY
  Total units: X,XXX | Days of supply: XX days
  ⚠ Low stock: [ASIN list]
  🚨 Urgent reorder: [ASIN list]

📢 ADVERTISING
  Ad spend: $X,XXX | ACOS: XX% | ROAS: X.Xx
  Best performer: [term] at X% ACOS
  Worst performer: [term] at X% ACOS — ACTION NEEDED

⚠ ALERTS
  [List of any anomalies detected]

✅ TOP 3 ACTIONS THIS WEEK
  1. [Highest priority action]
  2. [Second priority]
  3. [Third priority]

Workspace

Creates ~/backend-reports/ containing:

  • business/ — business report analyses by period
  • inventory/ — inventory health tracking
  • advertising/ — ad performance archives
  • financial/ — financial summaries
  • digests/ — weekly management digests

Output Format

Every report analysis outputs:

  1. Performance Summary — key metrics vs. prior period with trend indicators
  2. Anomaly Detection — unusual patterns that require attention
  3. Inventory Status — stock health with specific reorder actions
  4. Ad Performance — ACOS summary with specific optimization actions
  5. Financial Health — margin and fee analysis
  6. Priority Action List — top 5 actions ranked by expected business impact
  7. Next Week Forecast — projected performance based on current trends
Usage Guidance
This skill appears coherent for analyzing exported Seller Central CSV/text reports and does not request credentials or install code. Before using: (1) avoid pasting raw sensitive or personally identifiable data—sanitize or anonymize reports if possible; (2) confirm where the agent's workspace stores saved reports and who can access them, since 'report save' implies persistence; (3) note the skill lists 'allowed-tools: Bash' though it contains no shell commands — consider restricting shell execution if you don't want the agent running commands in your environment; (4) if you expect the skill to fetch data directly from Seller Central (automatically), it currently will not do that without additional credentials or code. If any of these concerns matter to you, review the referenced GitHub repo before installing or only use with non-sensitive sample data.
Capability Analysis
Type: OpenClaw Skill Name: amazon-backend-report Version: 1.0.0 The skill bundle is a legitimate template for analyzing Amazon Seller Central reports. It provides detailed analytical frameworks for business, inventory, advertising, and financial data without any evidence of malicious intent, data exfiltration, or unauthorized execution. The use of the Bash tool is consistent with the stated purpose of organizing analysis files in the '~/backend-reports/' directory (SKILL.md).
Capability Assessment
Purpose & Capability
Name and description match the SKILL.md: the agent expects the user to paste CSV/text exports from Seller Central and performs offline analysis. It does not request Seller Central credentials or unrelated services, which is coherent for a paste-based analyzer.
Instruction Scope
Instructions ask the user to paste exported report data and provide goals, periods, and ASIN lists — all reasonable. Note: pasted reports may contain sensitive financial or PII data; the skill also defines a 'report save' command (saving to the agent workspace), so users should be aware their data may be stored by the agent. The SKILL.md does not instruct the agent to fetch data from external endpoints or read system files, which limits scope creep.
Install Mechanism
No install spec or code files are present; this is instruction-only, so nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. This matches the stated paste-based workflow and is proportionate.
Persistence & Privilege
always:false and normal autonomous invocation are set (expected). The only persistence hint is the 'report save <period>' command which implies writing to the agent workspace—this is normal but users should confirm how/where the workspace stores data and who can access it.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install amazon-backend-report
  3. After installation, invoke the skill by name or use /amazon-backend-report
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Amazon Backend Report Analyzer initial release. - Automatically analyzes Amazon Seller Central business, inventory, advertising, and financial reports to surface actionable insights. - Supports paste-in CSV/text report analysis with clear performance summaries and prioritized action items. - Offers commands for analyzing various report types, comparing time periods, generating weekly digests, and managing workspace archives. - Frameworks and benchmarks included for business, inventory, advertising, and financial report analysis. - Outputs include performance summaries, anomaly detection, inventory status, ad performance, financial health, action lists, and next week forecasts.
Metadata
Slug amazon-backend-report
Version 1.0.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Amazon Backend Report?

Amazon Seller Central backend data report auto-analysis agent. Automatically analyze business reports, inventory reports, advertising reports, and financial... It is an AI Agent Skill for Claude Code / OpenClaw, with 145 downloads so far.

How do I install Amazon Backend Report?

Run "/install amazon-backend-report" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Amazon Backend Report free?

Yes, Amazon Backend Report is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Amazon Backend Report support?

Amazon Backend Report is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Amazon Backend Report?

It is built and maintained by mguozhen (@mguozhen); the current version is v1.0.0.

💬 Comments