Amazon Skills by Nexscope: 53 markdown playbooks you install with one npx command

Installing the collection is one command and the flag is what matters

The whole collection installs in a single line:

npx skills add nexscope-ai/Amazon-Skills -g

The trailing `-g` is the part to notice, because it is the difference between a skill you can use in this project and one your agent picks up everywhere. Drop it and the skills land in the current directory only.

If 53 folders is more than you want, the same tool takes a single skill by name:

npx skills add nexscope-ai/Amazon-Skills --skill amazon-keyword-research -g

Every skill name in the collection follows that pattern, prefix and hyphenation included, so the same flag works for `amazon-fba-calculator` or `amazon-ppc-campaign`. After installation there is no build step and no service to start. You open your agent and ask a question in ordinary language, for example asking for search volume, competition and an opportunity score for a keyword on Amazon US. The named agents in the compatibility list are OpenClaw, Claude Code, Cursor, Windsurf and Codex, plus anything that speaks the Skills format.

A skill is a markdown file, so your agent supplies the data

The stated design is deliberately plain: an Amazon skill is a text instruction file that gives an agent selling expertise, with no binaries, no API keys and no setup steps, just markdown that any model can read. That has two consequences worth holding onto before you install.

The first is that the method travels and the numbers do not. A keyword research skill covers long-tail mining from Amazon autocomplete, competition analysis, seasonality trends and opportunity scoring, but it cannot query Amazon itself. The output quality depends on what your agent already knows, what it can browse, and what you paste into the conversation. Read the result as a framework applied to inputs, not as a market report.

The second is that a skill is editable. Because each one is a file in the repository, you can open `amazon-fba-calculator`, change the assumptions it uses, and have every agent that installs it pick up your version. For a seller with house rules about fees or margins, that is the feature. The repository also lists Python as its primary language, so some of these folders carry more than prose.

The collection is grouped by job, and one section is short a skill

Skills are sorted into working groups rather than by marketplace. Product research and keywords holds five skills, including `amazon-trending-products` for BSR patterns and emerging niches, `amazon-niche-finder` for demand, competition, margin and growth scoring, and `amazon-seller-analytics` for storefront revenue estimation and portfolio shape. Pricing and profitability is the largest visible group at seven, covering the FBA calculator, `tariff-calculator-amazon`, `amazon-profit-analyzer`, `amazon-repricing-strategy`, `amazon-buy-box`, `amazon-deal-finder` and `amazon-shipping-calculator`. Advertising holds four, and competitor analysis is headed as five while showing four rows.

Each row carries a status: available means production-ready, beta means functional and still being improved. Reading those markers before installing is the fastest way to know which of the files you are willing to trust with a real pricing decision.

The repository root shows the rest of the catalogue even where the listing is cut off: content and creative work such as `amazon-a-plus-content`, `amazon-listing-images`, `amazon-product-photography` and `amazon-storefront-design`; account and compliance topics such as `amazon-suspension-appeal`, `amazon-category-ungating` and `amazon-product-compliance`; tracking with `amazon-keyword-tracker`, `amazon-rank-tracker` and `amazon-price-tracker`; cross-border work with `amazon-global-selling` and `amazon-international-listings`; and programs like `amazon-subscribe-save`, `amazon-vine-program` and `amazon-wholesale-sourcing`.

The money skills are calculators with named line items

The profitability group is the part with the most checkable arithmetic. `amazon-fba-calculator` breaks a fee into referral fees, fulfillment fees and storage costs, then works back to a net margin. `amazon-shipping-calculator` handles FBA and FBM shipping and fulfillment cost, including dimensional weight, storage and removals. `amazon-profit-analyzer` goes further with a revenue waterfall, hidden fees, ACoS impact and return costs.

Trade costs have their own file. `tariff-calculator-amazon` is billed as a universal calculator covering import duties, landed costs and VAT or GST for any trade route, which is the piece most relevant if you are moving inventory across a border rather than reselling inside one marketplace.

Two more turn rankings into estimates and rankings into plans. `amazon-sales-estimator` estimates monthly sales from BSR, an ASIN or a keyword and is described as working across any Amazon category. `amazon-deal-finder` plans Lightning Deals, Best Deals, Coupons and Prime Exclusive promotions with an ROI calculator attached. Again, the arithmetic is the deliverable: the inputs are yours to supply and to verify.

Advertising is four skills that reduce to targets and budgets

`amazon-ppc-campaign` builds campaign structures or works on existing ones, and its named jobs are calculating ACoS targets, grouping keywords and setting bid strategies. `amazon-advertising-strategy` sits above it and covers the three campaign types SP, SB and SD with budget allocation and ACoS optimization. `amazon-display-ads` handles Sponsored Display work: audience targeting, retargeting and creative optimization. `amazon-negative-keywords` is the waste-reduction file, doing search term analysis and estimating savings.

Two directories in the root belong to the same family even though the visible listing stops earlier: `amazon-dayparting-strategy` and `amazon-coupon-strategy`. Add `amazon-backend-keywords` and `amazon-search-optimization` from the keyword side, and the advertising work is largely about deciding where money goes and which queries to stop paying for.

`amazon-brand-analytics` is the one file in the group that reads a report rather than a campaign. It interprets Brand Analytics output, covering SFR analysis, Market Basket insights and Item Comparison positioning, which is a different skill from the rest and worth installing on its own.

Twelve marketplaces, one country list, and a US-first example

Coverage is stated as twelve marketplaces: US, UK, DE, FR, IT, ES, JP, CA, AU, IN, MX and BR. Keyword research is the skill that repeats that number in its own row, so it is the one to trust for cross-market coverage. Most of the rest of the collection does not name a marketplace at all, which leaves you to check whether the fee assumptions in the calculator files match the marketplace you actually sell in.

The sample prompt the collection ships with is US specific: research a keyword on Amazon US and report search volume, competition and an opportunity score. If you sell in Germany or Japan, that example is a template rather than a promise, and the currency, fee schedule and VAT treatment all change per marketplace.

There is a sibling collection too. The eCommerce-Skills repository is advertised as 142 skills across Amazon, Shopify, eBay, Etsy, TikTok Shop and Walmart, so this one is the Amazon slice of a larger set rather than the whole catalogue.

Every link routes to the vendor's own skill hub

The repository is MIT licensed and free, and every skill name in it links to `nexscope.ai/skillhub` with a `?co-from=github` parameter appended. The Discord invite sits in the same header. Nothing about that makes the files less useful, but it is worth knowing that the free repository is also the top of the vendor's funnel, and that clicking a skill name leaves GitHub for the company's own pages.

There is a version story to keep straight too. The repository publishes no GitHub releases at all, so there is no tag history to check and nothing to pin. The last push to the default branch was 2026-08-26, which is recent enough that the collection is moving, and no individual skill carries a version number of its own.

The project is small in open-issue terms: 728 stars, 126 forks and 3 open issues when this snapshot was taken. For a collection this size that ratio suggests the skills get installed and rarely argued with, which is also why the status markers in the listing matter more than the issue count.

Editorial conclusion

Amazon Skills earns a place if you sell on Amazon and want a written method rather than a data feed: the fee math, the keyword framework and the PPC structure are all there as readable files you can edit. Skip it if you expected live Amazon data, because there are no binaries and no API keys in the box, and check the listing before you install, since the root carries 52 skill folders against a headline count of 53.

Frequently asked questions

Official sources

1. Issues
2. License: MIT
3. nexscope-ai/Amazon-Skills on GitHub
4. README