Grailed Real-Time Data

Live resale market intelligence from Grailed — thrift price comps, closet inventory, seller stats, and a vintage deal finder — as clean structured JSON.

Built for vintage & thrift resellers (Poshmark, eBay, Depop, Mercari cross-listers), consignment shop owners, resale arbitrageurs, fashion market researchers, pricing intelligence teams, and AI agents. Enable only the features you need, click Start, and stream results to your dataset, webhook, or LLM pipeline.

⚠️ Free-tier notice: Free Apify accounts get a 2-result trial per run. Upgrade to any paid Apify plan for unlimited exports. See Free tier limits.

---

Why teams use this Actor

	Grailed Real-Time Data	Typical browser scraper
Speed	Fast per-item collection	Often seconds per page
Memory	512 MB default, streams row-by-row	2–4 GB+
Setup	Checkbox UI, prefilled test input	Fragile & high-maintenance
Cost	Low compute	High
Output	Structured JSON, LLM-ready	Often messy HTML
Scale	10,000+ item runs without ballooning RAM	Memory-heavy buffers
Delivery	Dataset + optional real-time webhook	Export-only

---

Feature matrix

Feature	Default	What it does
🔍 Listing Search	ON	Live keyword search for thrift brands, streetwear, and vintage finds with brand, size, condition, and min/max price filters
✨ Enrich with full listing details	off	Adds description, fabric/material tags, style tags, full image gallery, and seller shipping discounts to the same search row
📦 Listing Details	off	Full records for specific listing IDs or Grailed URLs
👚 Closet Listings	off	Full active inventory of any reseller/thrift closet — track competitor stock
👤 Seller Profile	off	Reseller stats: love notes, items sold, closet size, verification, bio
🧾 Sold Item Comps / History	off	Completed and sold listings to calculate real market comps, resale margins, and sell-through rates
🔗 Scrape By URL	off	Paste any Grailed URL — listings, search URLs, closets, category pages, brand feeds

All defaults are set to 10 results so your first run finishes in seconds. Raise the caps once you've confirmed the output shape.

---

Use cases

• Finding undervalued thrift & vintage inventory — hunt vintage single-stitch tees, designer denim, gorpcore, and Y2K with price ceilings (e.g. vintage 90s levis 501 under $40).
• Pricing comps & sell-through analysis — pull completed and sold listings for any keyword or seller to compute real market comps, average sold prices, and resale margins before you buy.
• Closet inventory monitoring — watch top competitors' and consignment sellers' active stock; get alerted the moment something new drops.
• Instant Slack/Discord deal alerts — pipe every new row through a webhook to Slack, Discord, Zapier, Make, n8n, or your own pricing bot for newly listed steals.
• Multi-platform pricing sync & inventory feeds — one JSON schema feeds your cross-listing, repricing, or catalog pipelines.
• AI & agent workflows — structured JSON ready for Claude, ChatGPT, LangChain, LlamaIndex, RAG, scoring, and alerts.

---

What you get

Every record includes featureType, url, and scrapedAt so you can filter, join, and pipe into any workflow.

Listing Search (featureType: "listing_search") — one product = one row

Field	Description
listingId, title, url	Identity and listing link
price, originalPrice, discountPercent, priceDropped	Pricing and markdown signals
currency, shippingCost, freeShipping	Cost context
brand, designers[], size, condition, color	Attributes
category, subCategory, department, marketTier	Catalog placement (grails / hype / sartorial / basic)
likes, views, numberOfOffers	Demand signals
sellerUsername, sellerLocation, sellerVerified	Seller context
imageUrl, imageUrls[]	Cover photo and gallery
postedAt, timeSinceListed, isNewlyListed	Freshness — critical for deal sniping
position, keyword	Search context
detailsFetched	false = search card; true = full fields merged into this same row

With "Enrich with full listing details" on, the Actor does not create a second row and never drops partial items. It enriches the search row in place with:

Extra fields when detailsFetched: true
description	Full listing text
materialTags, material	Fabric / material tags
styleTags, hashtags, tags	Style taxonomy
imageUrls[]	Full image gallery
sellerShippingDiscounts	Seller shipping-discount signals
measurements[], itemSpecifics[]	Garment measurements and specifics

Example: searchMaxResults: 10 + enrichment on → 10 dataset rows, not 20, and nothing filtered out.

Listing Details (featureType: "listing_details")

Full product rows only when you enable the Listing Details feature and pass listing IDs / URLs. Search enrichment merges into listing_search instead.

Closet Listings (featureType: "closet_listings")

One row per active listing in a seller's closet — price, brand, size, condition, photos, and freshness.

Seller Profile (featureType: "seller_profile")

Username, display name, bio, verification, rating, reviews, items sold, closet size, response rate, and recent love notes (feedback) when available.

Sold Item Comps (featureType: "sold_comps")

Completed and sold listings by seller or by keyword — sold price, sold date, brand, size — the raw material for market comps and margin math.

Scrape By URL (featureType: "scrape_by_url")

Paste any Grailed URL. The Actor auto-detects the page type (listing, search, seller, brand, category) and returns structured rows.

---

Full input reference

Enable only what you need. Listing Search is on by default with thrift-ready sample keywords.

There is no separate global cap — the run's total volume is simply the sum of the section caps you set (keywords × max per keyword, sellers × max per closet, etc.), so you control everything from the feature sections.

Free plan: a 2-result per-run trial cap applies on top of your section caps — see Free tier limits.

Input	Type	Default	Description
Listing Search
enableListingSearch	boolean	true	Live keyword search
searchKeywords	string[]	carhartt / levis / patagonia samples	Search terms
searchMaxResults	integer	10	Max per keyword (1–200)
searchDepartment	enum	any	menswear, womenswear
searchDesigner	string	—	Brand filter (e.g. Carhartt, Levi's, Patagonia, The North Face, Stussy)
searchSize	string	—	e.g. M, L, XL, 32, 34
searchCondition	enum	any	new_with_tags, new_without_tags, gently_used, well_worn
searchMinPrice / searchMaxPrice	integer	—	USD price bounds — set a max for deal hunting
searchStrata	enum	any	grails, hype, sartorial, basic
searchSort	enum	relevance	relevance, newest, price-low, price-high, heat
searchSoldOnly	boolean	false	Search completed/sold items instead of live inventory
searchFetchFullDetails	boolean	false	Enrich each search row in place (same row, detailsFetched: true)
Listing Details
enableListingDetails	boolean	false	Full product rows
listingUrls	string[]	—	Listing URLs or bare numeric IDs
Sellers
enableClosetListings	boolean	false	Closet inventory
enableSellerProfile	boolean	false	Profile + love notes
enableSoldComps	boolean	false	Sold comps by seller and/or keyword
sellerUsernames	string[]	—	Without @
sellerMaxListings	integer	30	Cap per closet (1–500)
soldMaxItems	integer	30	Cap sold items per seller/keyword (1–200)
soldKeywords	string[]	—	Direct sold-market comp searches (e.g. carhartt j97, stussy 8 ball)
URL
enableScrapeByUrl	boolean	false	Any Grailed URL
scrapeUrls	string[]	—	Listing, search, seller, category, or brand URLs
scrapeMaxPages	integer	3	Depth for list-style URLs (1–20)
Options & delivery
includeRaw	boolean	false	Extended structured block per row (larger output)
webhookUrl	string	—	Optional real-time POST per row
webhookFormat	enum	json	json (full record) or slack
proxyConfiguration	object	Residential US	Recommended — see Proxy

Practical thrift search examples

Goal	Keywords + filters
Vintage denim flips	vintage 90s levis 501, levis 517 made in usa + searchMaxPrice: 60
Workwear goldmines	carhartt detroit jacket, carhartt double knee + condition: gently_used
Gorpcore sourcing	patagonia fleece, arc'teryx gamma + searchSort: price-low
Streetwear steals	stussy 8 ball, supreme box logo + searchMaxPrice: 120
Sold comps before you buy	Sold comps with soldKeywords: ["carhartt j97"]
Y2K restock alerts	vintage y2k + webhook to Discord + searchSort: newest

---

Output & streaming

• Every row is written to the dataset the moment it is ready — long runs never hold the full result set in memory, so RAM stays flat even at 10,000+ items.
• Parallel workers process multiple keywords, sellers, details, and URLs concurrently with randomized pacing between chunks.
• Spending limits — set a max cost per run in the Console and the Actor stops gracefully with a summary when the limit is reached. No crashed runs, no overspend.
• Filter the dataset by featureType.
• Dataset views: overview, search, listing_details, closet_listings, seller_profile, sold_comps, scrape_by_url.
• Each run writes a summary object (counts per feature, paywall status) to the run's OUTPUT.

Row counts to expect

What you enable	Dataset rows
Listing Search, 10 results, enrichment off	10 × listing_search (detailsFetched: false)
Listing Search, 10 results, enrichment on	10 × listing_search (detailsFetched: true, richer fields on the same row)
Listing Details with 5 URLs	5 × listing_details
Closet Listings, max 30	Up to 30 × closet_listings per seller
Sold comps, max 30 per keyword	Up to 30 × sold_comps per keyword

Sample search row (enriched)

{
  "featureType": "listing_search",
  "listingId": "5123456",
  "title": "Vintage Carhartt Detroit Jacket Brown Duck Canvas",
  "price": 85,
  "originalPrice": 120,
  "discountPercent": 29,
  "currency": "USD",
  "brand": "Carhartt",
  "size": "L",
  "condition": "Gently Used",
  "marketTier": "basic",
  "likes": 34,
  "sellerUsername": "vintagevault",
  "detailsFetched": true,
  "description": "90s Carhartt J97 Detroit jacket…",
  "materialTags": ["cotton duck", "blanket lining"],
  "styleTags": ["workwear", "vintage"],
  "imageUrl": "https://…/cover.jpg",
  "imageUrls": ["https://…/1.jpg", "https://…/2.jpg"],
  "postedAt": "2026-09-20T14:03:00.000Z",
  "timeSinceListed": "6d 2h",
  "url": "https://www.grailed.com/listings/5123456",
  "scrapedAt": "2026-09-26T12:00:00.000Z"
}

---

Webhook integration

Every record is always saved to the dataset. Optionally, each new row is also POSTed in real time to webhookUrl — perfect for instant deal alerts on rare thrift finds or low-priced comps.

JSON format

Full record object, identical to the dataset row. Point it at Discord, Zapier, Make, n8n, or your own pricing bot.

Slack format

A formatted message:

:shirt: *Vintage Carhartt Detroit Jacket Brown Duck Canvas*
*Type:* listing_search  •  *Price:* USD 85  •  *Brand:* Carhartt  •  *Size:* L
*Condition:* Gently Used  •  *Likes:* 34  •  *Seller:* vintagevault
<https://www.grailed.com/listings/5123456|Open on Grailed>

Setup: paste a Slack Incoming Webhook URL (or a Discord webhook using the /slack style URL) into Webhook URL and pick Slack message as the format. Webhook failures never interrupt the run or dataset writes.

---

MCP / AI Agent usage

Output is structured JSON — ready for Claude Desktop, Cursor, LangChain, LlamaIndex, and custom agents via the Apify MCP server:

1. Add the Apify MCP server to Claude Desktop / Cursor.
2. Expose this Actor to the MCP server.
3. Ask questions in natural language.

User: "What is the average sold price for a 90s Carhartt J97 jacket on Grailed?"
→ Agent runs the Actor with enableSoldComps=true, soldKeywords=["carhartt j97"], soldMaxItems=50
→ Agent reads the dataset
→ Agent answers: "Average sold price is $142 across 50 comps; median $120; sell-through concentrated in sizes L–XL."

Other agent-friendly prompts:

• "Find 10 Patagonia fleece listings under $30 and rank by likes per dollar."
• "Monitor @thriftgod's closet daily and alert me on new Stussy items."
• "Compare sold comps for vintage 501s across the last 50 sales."

API quick start

curl -X POST "https://api.apify.com/v2/acts/YOUR_ACTOR_ID/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "enableListingSearch": true,
    "searchKeywords": ["carhartt detroit jacket"],
    "searchMaxResults": 10
  }'

Dataset items: GET https://api.apify.com/v2/datasets/{datasetId}/items?format=json

---

Proxy

Residential proxies recommended. The Actor defaults to Apify Residential with a US exit, which you can change in the Proxy settings input (US, CA, GB, DE, AU, …). Match the country to where you want the marketplace to place you.

• Free-tier runs and paid runs use the same proxy configuration.
• Rotating residential sessions are used automatically for reliability at scale.
• Long runs (the Actor supports runs up to ~2h 46m of configured timeout) benefit from leaving the default settings untouched.

---

Free tier limits

Apify injects plan signals into every run; this Actor uses them to keep the free funnel alive while reserving unlimited exports for paid plans:

Plan	Behavior
Any paid plan (Bronze → Diamond)	Normal run, full output, no caps — logged as Paying user — full output.
Free plan	Results capped at 2 items per run (default), with a clear upgrade message.
Free plan (block mode)	Owners can set FREE_TIER_MODE=block so the Actor exports 0 items with a friendly message instead.

Notes:

• The cap is transparent — it is stated here, in the input schema, and in the run logs; it is never presented as an error.
• Runs always exit gracefully; a capped run is a successful run.
• Owner-configurable via Console environment variables: FREE_TIER_MODE (limit default | block) and FREE_TIER_MAX_ITEMS (default 2).

---

FAQ

Do I need code? No — use the Console UI. Developers can use the Apify API, schedules, and webhooks.

Which proxy should I use? Residential, matching your target region. The default (Apify Residential, US) works for most thrift hunting.

How fresh is the data? Records are collected live at run time — schedule runs as often as your plan allows for monitoring and alerts.

How do I avoid rate limiting? Keep per-keyword caps reasonable, let the built-in randomized pacing do its job, and use residential proxies. For very large inventories, split runs by seller or keyword.

What happens when I hit my spending limit? The Actor stops collecting, writes the summary, and exits cleanly — no partial or corrupted exports.

Multiple features in one run? Yes — every row is tagged with featureType, so one run can search, enrich, pull closets, profiles, and sold comps together.

Large runs? Rows stream continuously and work runs in parallel; memory stays flat even on 10,000+ item runs.

Is the 2-item free cap really all I get? Yes — upgrade to any paid Apify plan to lift it. The cap is enforced transparently and never appears as an error.

LLM / agents? Yes — structured JSON + Apify MCP for agent-driven runs. See the MCP section above.

---

Contact / custom projects

Need something tailored? I build custom scrapers, data pipelines, and web apps of any kind — marketplaces, lead gen, internal tools, dashboards, and full-stack products.

• Email: dubem115@gmail.com
• GitHub: https://github.com/DrunkCodes

Open to project work, integrations, and ongoing data infrastructure.