Reviews, listings, costs, growth and inventory
Ecommerce Skills
From Customer Evidence to Store Decisions
Nexscope’s collection organizes ecommerce work into separate skills. Bring real reviews, product details, costs or inventory records, then choose a workflow for a traceable analysis or action draft. The depth of guidance and supporting code varies substantially between folders.
Hosts listed by the author
- Claude Code
- Codex
- Cursor
- Windsurf
- OpenClaw
- GitHub stars
- 1,291
- Checked
Start with something you can check against the source
For a first run, analyze one SKU’s reviews and verify the themes and quotations before changing its listing. If operating data is incomplete, resolve the cost or inventory definitions first. A polished report should not conceal missing facts.
Pick a task and get started
The default command installs only review analysis. For another task, use its displayed command, then attach the relevant files and send the prompt in a session where the skill is loaded.
Use a real review sample to separate complaints, praise and feature requests. Preserve its scope: themes alone do not prove a defect or establish review authenticity.
Install one skill for the task
Run the author’s single-skill command in your local terminal; -g installs at user level. Select your assistant when prompted and confirm the skill loads. Use the corresponding command for another task. The command retrieves repository content; this guide was reviewed at the pinned commit below.
Read the author’s setup guidenpx skills add nexscope-ai/eCommerce-Skills --skill product-review-analysis -gFirst run: turn a review file into a checkable report
Prepare reviews.csv for one SKU with review_id, date, rating, variant, review text and source. Put the platform, market and reporting period in product-context.md. After installation, confirm product-review-analysis appears in the host’s available skills and reload the session if required. Attach both files and use the first task prompt. Open reports/review-analysis.md, spot-check quotations and IDs, and recalculate theme denominators before acting.
Read the author’s single-skill installation instructionsGive your AI this first task
These are editorial invocation examples based on the reviewed source. Filenames are input conventions to replace with your own materials. Expected deliverables are not results from a run by this site; leave unsupported fields unknown.
Use product-review-analysis with reviews.csv and product-context.md for this SKU. Establish the platform, period, variants, valid sample count and duplicates first; identify missing files. Separate purchase drivers, recurring complaints and feature requests. For each theme, report its count, denominator, representative quotation and review_id. Allow multiple themes per review and explain overlapping percentages. Compare periods or variants only with adequate samples. Propose three product, FAQ or listing improvements, each with evidence, an explanation to test and a way to check the outcome. Do not invent authenticity percentages, causal rating effects or revenue gains. Omit cross-brand comparisons without competitor evidence. Save reports/review-analysis.md, spot-check quotes and counts, and explain which buyers the sample cannot represent.
Check what you receive
- A theme table traceable to review IDs, separating observations, interpretations and proposals.
- Three supported improvements with validation steps; verify quotations, counts and denominators instead of filling unsupported template fields.
More about this skill
The collection divides ecommerce work into individual skills for competition, pricing, growth, marketing, listings, advertising, monitoring, supply chains and platform topics. This page develops five paths with identifiable inputs and outputs. They can connect: reviews reveal customer concerns, listing work improves how the product is explained, costs constrain promotions, growth planning prioritizes experiments, and inventory analysis tests fulfillment readiness.
Each SKILL.md gives the assistant a method, but folders are not equivalent. Some contain a framework, some call for web tools, and others include Python code. The homepage’s description of an entirely Markdown collection is incomplete. The selected review, listing, growth and warehouse workflows mainly rely on the assistant. The Shopify margin folder includes a calculator, but its implementation still models Amazon costs.
Guidance varies in depth too. Listing generation has separate Create and Optimize modes, a keyword table and before/after output. Growth planning uses five growth levers, opportunity prioritization and a 90-day roadmap. By comparison, the Beta shopify-email-flows document gives broad planning guidance. Available and Beta are the author’s labels, not performance tests conducted by this site.
The repository uses the MIT license; model and external-service costs are separate. Nexscope promotes monitoring and market-data services, but installing a skill neither provides those services nor makes them your only data option. Your exports, authorized data tools and public sources can supply evidence. Coverage, data quality and business outcomes each need their own checks.
Sources and verification
Sources checked
The selected documents and implementation were read at the pinned commit, with installation syntax, five workflows and example attribution checked. No skills were installed, third-party scripts run or stores accessed.