Skip to content
Back to skills

macys-research

ASecurity

Looks up a Macy's product's full detail and customer reviews by its numeric productId, and pulls Macy's own search-box (typeahead) suggestions, using the Crawlora API, returning clean JSON. Use when the user has (or can get) a Macy's product page ?ID= value and wants its price/description/variants/reviews, or wants Macy's autocomplete suggestions for a query — instead of scraping macys.com. There is no full-text product search or category browse — a known numeric productId is required as the ...

  • 0 votes
  • 0 copies
  • 7 views
  • Added August 14, 2026
ai-agentsgogitapisecurity

Works with

  • api

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned August 14, 2026

npx -y skills add Crawlora-org/crawlora-skills --skill macys-research --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of macys-research?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for macys-research
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/crawlora-org-macys-research/badge)](https://www.skillsdirectory.com/skills/crawlora-org-macys-research)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: macys-research
description: Looks up a Macy's product's full detail and customer reviews by its numeric productId, and pulls Macy's own search-box (typeahead) suggestions, using the Crawlora API, returning clean JSON. Use when the user has (or can get) a Macy's product page ?ID= value and wants its price/description/variants/reviews, or wants Macy's autocomplete suggestions for a query — instead of scraping macys.com. There is no full-text product search or category browse — a known numeric productId is required as the entry point.
---

# Macy's research

Look up a specific Macy's product's detail and reviews by its numeric
productId, and pull Macy's own typeahead suggestions — all as normalized
JSON from the Crawlora API, with no HTML scraping. This skill does not
cover search or category browsing; it only works from a known productId.

## When to use this skill

- "What does this Macy's product cost / what's in the description?" given
  a macys.com product URL or its `?ID=` value.
- "Pull the reviews and rating summary for this Macy's product."
- "What does Macy's autocomplete suggest for [query]?" to discover related
  search terms.
- "Summarize this Macy's item's colors, pricing, and customer sentiment"
  for a known product.

## Setup (one-time)

- Get a free Crawlora API key (2,000 credits/mo, no card) at [https://crawlora.net](https://crawlora.net?utm_source=github&utm_medium=referral&utm_campaign=crawlora-skills).
- `export CRAWLORA_API_KEY=sk_your_key_here`
- All requests: `x-api-key: $CRAWLORA_API_KEY` against
  `https://api.crawlora.net/api/v1`. Missing/invalid key → `401`.

## How it works

1. **Get a productId** — a numeric id taken from a macys.com product page's
   `?ID=` query parameter (or otherwise already known/discovered). There is
   no search or category endpoint in this skill to find one from a keyword.
2. **Detail** — `/macys/product/{productId}` returns full product content:
   name, brand, description, department/division, category breadcrumb,
   pricing (with sale detection), availability, images, aggregate rating,
   and every purchasable color variant with its own price.
3. **Reviews** — `/macys/product/reviews` (`product_id`, paginated with
   `page`) returns one page of normalized customer reviews plus a
   site-wide rating summary (count, average, recommended ratio, histogram)
   for that product.
4. **Suggest** — `/macys/suggest` (`query`) returns Macy's own search-box
   typeahead phrases for a partial query — a flat list of strings, no
   product data — useful for discovering related terms, not for finding a
   productId.

Full endpoint list, methods, and params: [`reference/endpoints.md`](reference/endpoints.md).

## Calling the API

```sh
# Product detail (productId is a path param):
scripts/crawlora.sh /macys/product/12345678 | jq '{name,price,rating}'

# Reviews for that same product, paginated:
scripts/crawlora.sh /macys/product/reviews product_id=12345678 page=1 | jq '.'

# Search-box suggestions for a partial query:
scripts/crawlora.sh /macys/suggest query="winter coat" | jq '.'
```

Raw `curl` fallback:

```sh
curl -fsS -H "x-api-key: $CRAWLORA_API_KEY" \
  "https://api.crawlora.net/api/v1/macys/product/12345678" | jq '.'
```

## Endpoint reference

See [`reference/endpoints.md`](reference/endpoints.md) for every Macy's
endpoint this skill uses (method, path, params, description).

## Examples

- **Product due diligence:** `/macys/product/{productId}` for
  price/description/variants, then `/macys/product/reviews` (same
  `product_id`, paginated) to summarize what customers say before
  recommending it.
- **Term discovery:** `/macys/suggest` on a rough query to see Macy's own
  suggested phrases — useful context when a user asks about a product
  category, even though this skill can't browse or search that category
  directly.

## Notes & limits

- **No search or category-browse endpoint** — this skill requires a known
  numeric `productId`, sourced from a macys.com product page's `?ID=`
  query parameter, as the entry point. `/macys/suggest` only returns
  suggested search phrases (no product data), so it cannot substitute for
  search when the user doesn't already have a productId.
- **Credits / pay-on-success:** billed only on `2xx`; free tier 2,000 credits/mo.
  Key at [https://crawlora.net](https://crawlora.net?utm_source=github&utm_medium=referral&utm_campaign=crawlora-skills).
- **Public data only** — public Macy's product pages; respect Macy's terms.
- **Security:** key lives in `CRAWLORA_API_KEY` only — never hardcode, query-param, or commit it.
- `product_id` on `/macys/product/reviews` is the same numeric id as
  `productId` on `/macys/product/{productId}` — a product with zero
  reviews, or a well-formed but unrecognized id, returns a normal, empty
  result rather than an error.
- Reviews are paginated with a 1-based `page`, defaulting to 1.

Files in this skill

  • SKILL.md4.8 KB
  • reference/endpoints.md2.3 KB
  • scripts/crawlora.sh1.8 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.