How to Scrape Mercari in 2026 (API & Python)

Mercari doesn't run a public developer portal for reading its catalog, so the fastest reliable path in 2026 is a structured scraping API that returns normalized JSON for search results, item detail, and taxonomy — instead of parsing Mercari's listing pages yourself. Mercari sits in the same resale genre as StockX and Poshmark, but with a wrinkle those don't have: Mercari US and Mercari Japan are two different marketplaces under one brand, with separate apps, separate item catalogs, and separate currencies — this guide covers the US site.

Why scrape Mercari data?

• Resale-price research — track what an item, brand, or category actually sells for on a large peer-to-peer marketplace, as a comp against retail or other resale sites.
• Cross-market arbitrage research — compare US listing prices against another marketplace's asking prices for the same brand or item type (note: Mercari Japan is a separate catalog, not the same inventory).
• Brand and category trend tracking — pull listing counts and pricing for a brand or category to see what's circulating and at what price band.
• Market research — use the category/brand/size taxonomy to size how many active listings exist in a given niche without guessing spellings.
• AI pipelines and agents — feed normalized listing and item data into a pricing or sourcing agent instead of screen-scraping HTML.

Is it legal to scrape Mercari?

Option 1: DIY in Python (and why it breaks)

A naive scraper fetches a search results page and tries to pull listing cards out of the rendered markup:

import requests
from bs4 import BeautifulSoup

resp = requests.get(
    "https://www.mercari.com/search/?keyword=nike+jacket",
    headers={"User-Agent": "Mozilla/5.0 (compatible; research-bot/1.0)"},
)
soup = BeautifulSoup(resp.text, "html.parser")
# Listing cards render from client-side state (React/Next.js), so a plain
# HTML parse mostly returns an empty shell — you end up reverse-engineering
# the app's internal JSON endpoints instead

It demos once, then breaks:

• The Prohibited Conduct policy is explicit. It names robots, spiders, crawlers, and scrapers directly as a violation, separate from any technical block.
• Anti-bot defenses. Mercari fronts its site with bot detection; datacenter IPs and headless requests get rate-limited or blocked quickly, especially on search.
• Client-rendered markup. Search results and item pages render from JavaScript state, not stable server HTML — a layout change silently breaks CSS-selector scraping.
• No sanctioned alternative. Unlike eBay, Mercari doesn't run a developer portal at all, so there's no official-but-gated API to fall back on for catalog reads.

Option 2: No-code / ready-made tools

Browser extensions and point-and-click scrapers can export a single search page or item snapshot, but tracking a brand's resale prices or a category's listing volume over time means re-pulling the same pages on a schedule — a job for a stable endpoint with stored history, not a one-off export.

Option 3: A structured Mercari API

For a repeatable workflow, a Mercari scraping API returns normalized JSON for search, item detail, the home feed, autocomplete, and taxonomy — no page parsing, no app that doesn't exist. Search by keyword:

curl "https://api.crawlora.net/api/v1/mercari/search?query=nike+jacket" \
  -H "x-api-key: $CRAWLORA_API_KEY"

{
  "code": 200,
  "msg": "OK",
  "data": {
    "query": "nike jacket",
    "total_results": 21523,
    "items": [
      {
        "id": "m31455228584",
        "url": "https://www.mercari.com/us/item/m31455228584",
        "title": "Nike varsity jacket.   New with tags!",
        "thumbnail_url": "https://u-mercari-images.mercdn.net/photos/m31455228584_1.jpg?1784911712",
        "price_cents": 4873,
        "original_price_cents": 5400,
        "num_likes": 1931,
        "seller_id": "794269920",
        "created_at_unix": 1774057908,
        "updated_at_unix": 1784911712
      }
    ]
  }
}

query is required. Pull full item detail, autocomplete suggestions, or the taxonomy:

import requests

h = {"x-api-key": "YOUR_API_KEY"}
base = "https://api.crawlora.net/api/v1/mercari"

hits = requests.get(f"{base}/search", headers=h, params={"query": "nike jacket"}).json()["data"]["items"]
item_id = hits[0]["id"]

item = requests.get(f"{base}/item/{item_id}", headers=h).json()["data"]
suggestions = requests.get(f"{base}/autocomplete", headers=h, params={"query": "nik"}).json()["data"]
taxonomy = requests.get(f"{base}/master", headers=h).json()["data"]

Item detail carries the full photo set, price, and a similar_items list Mercari itself surfaces (real fields — check the docs):

{
  "code": 200,
  "msg": "OK",
  "data": {
    "id": "m38581776856",
    "url": "https://www.mercari.com/us/item/m38581776856",
    "title": "Women's NIKE WNBA Nike Mesh Jacket Size NEW MSRP 120",
    "description": "Women's NIKE WNBA Nike Mesh Jacket Size NEW MSRP 120",
    "photos": [
      "https://u-mercari-images.mercdn.net/photos/m38581776856_1.jpg?1777492193",
      "https://u-mercari-images.mercdn.net/photos/m38581776856_2.jpg?1777492193"
    ],
    "price_cents": 4578,
    "condition_code": 1,
    "category_code": 6,
    "shipping_from_state": "New Jersey",
    "similar_items": [
      {
        "id": "m34982910331",
        "url": "https://www.mercari.com/us/item/m34982910331",
        "title": "Lululemon Every Moment Pant *26\" Black Size 6",
        "thumbnail_url": "https://u-mercari-images.mercdn.net/photos/m34982910331_1.jpg?1783333561",
        "price_cents": 4000,
        "num_likes": 1967,
        "seller_id": "790985424",
        "created_at_unix": 1783333561
      }
    ]
  }
}

/mercari/home returns the same normalized shape as search but for Mercari's curated home feed, /mercari/autocomplete turns a partial query into suggested search terms, and /mercari/master returns the full category tree, brand list, and size list in one call so you can enrich category_code and condition_code values instead of guessing at Mercari's internal mapping.

What you can collect

• Search and home feed results: id, title, thumbnail, price in cents, original price when discounted, like count, seller id, and created/updated timestamps
• Item detail: full description, photo array, price, condition and category codes, shipping-from state, and a similar_items list
• Autocomplete: suggested search terms for a partial keyword, mirroring Mercari's own search-box suggestions
• Taxonomy: full category tree (with parent/child ids), brand list, and size list (with short names and codes)

Limitations

• No official API exists. Mercari doesn't run a developer portal at all — there's no gated-but-sanctioned alternative to fall back on, unlike eBay.
• The Prohibited Conduct policy is explicit. It names robots, spiders, and scrapers directly as a violation, independent of any technical block.
• US and Japan are separate marketplaces. These endpoints cover Mercari US (mercari.com) — item ids, categories, and prices don't map to Mercari Japan's (jp.mercari.com) separate catalog and yen pricing.
• Prices and like counts are a snapshot. Mercari items get repriced, liked, bundled, or marked sold constantly — timestamp what you store.
• Public data only. This covers public search, item, and taxonomy data — not private messages, offers, or a way to place an order on a buyer's behalf.

Where this gets used

• Resale and sourcing research — compare a brand's Mercari asking prices against retail or another resale marketplace.
• Brand and category sizing — measure listing volume and price bands for a brand or category using the taxonomy plus search.
• Pricing and sourcing pipelines — feed normalized item data into an agent that flags underpriced listings or tracks a watchlist.

Sources

Start collecting

Try it first, free: run any public URL through the Free Web Scraper, or check whether a site blocks bots with the Anti-Bot Checker — no signup.

Test the search, item, home, autocomplete, and master endpoints in the Playground, check the request and response schema in the API docs, and review credit costs on the pricing page. Mercari is a general secondhand marketplace, closer in shape to Poshmark's social-selling listings and Vinted's peer-to-peer fashion resale than to StockX's order-book pricing — pair how to scrape StockX alongside this guide for a fuller resale-market picture. See also is web scraping legal.

Part of our how-to-scrape guide series — every platform we cover, in one index.