Craigslist Scraper & API - Listings, Prices, Cars, Homes, Jobs

This Craigslist scraper and unofficial Craigslist API searches any Craigslist site (US, Canada, UK, Europe, Asia, worldwide) by keyword, area, category, price, postal code + radius and the site's own filters, and exports the listings to JSON, CSV or Excel. Use it as a Craigslist price tracker, a Craigslist search alert, for used car and for-sale prices, rental listings, local job ads, resale and flipping research, and market data for AI agents.

• What goes in: search keywords, Craigslist areas (sfbay, newyork, london, or a whole country like US), categories (sss for sale, cta cars & trucks, apa apartments, jjj jobs ...), price range, ZIP / postal code + miles, "has image", "posted today", titles-only, owner / dealer, sort, or Craigslist search / posting URLs.
• What comes out: one row per listing: title, price, currency, place, neighborhood, approximate map position, posted date, category, all photos, URL, plus bedrooms / sqft (housing), odometer (cars), salary / job title / company (jobs). Optional: full description, all attributes (condition, make / model, year, VIN, title status, size, bathrooms, pets, laundry ...), updated date.
• Typical cost: $0.45 per 1,000 listings, details included, no start fee. A 20-listing check costs under 1 cent.
• When to use it: pricing a used item or car before you buy or sell, watching a search for new listings, comparing rents across neighborhoods, collecting local job ads, finding underpriced items to flip, feeding live Craigslist data to an AI agent.

It reads the same JSON the craigslist.org search page loads, over plain HTTP, with no headless browser and no login: one request returns up to 360 listings with titles, and a full search list of up to 10,000 postings comes in about 10 requests. Deep search goes past Craigslist's 10,000-per-search limit automatically: in our tests all of "furniture" in the SF bay area returned 22,027 of 22,027 listings in 29 seconds (31 requests), and everything for sale in the East Bay returned 82,877 of the 83,372 the site counted (99.4%) in 74 seconds (138 requests). Unofficial: not affiliated with or endorsed by craigslist.

Sample output

Real rows from a cloud run for bike in the SF bay area (sfbay):

title	priceText	location	category	postedAt
Giant Revolt Advanced 1 Gravel Bike	$1,400	napa county	bik	2026-10-20T00:25:39Z
Juiced Ripcurrent S XL 32+mph Fat Tire ebike	$1,600	brentwood / oakley	bik	2026-10-20T00:25:56Z
Shimano Dura-Ace FC-R9200 Crankset 165MM 52/36	$525	san jose north	bop	2026-10-20T00:28:29Z
Harley 2006 Dyna Street Bob FXDB **AS NEW 900 Original Miles!!	$12,400	morgan hill	mcy	2026-10-20T00:21:00Z

Use cases

• Price research: what a used car, bike, phone or couch really sells for in your area, from hundreds of live listings with price, odometer, year and condition.
• New-listing alerts: schedule sort: "newest" every hour and send only new ids to Slack, Telegram or e-mail.
• Flipping and resale: find underpriced items across many Craigslist sites (or a whole country) in one run.
• Rental market data: compare rents, bedrooms and sqft across neighborhoods with apa (apartments).
• Local job ads: collect jjj postings with job title, salary, employment type and company.
• AI agents: {"searchQueries": ["iphone 13"], "areas": ["chicago"], "maxResults": 20} returns live listings for under 1 cent.

Pricing

$0.45 per 1,000 listings (pay per result). Posting details (description, all attributes, VIN, odometer, bedrooms, salary) are included at the same price, and there is no start fee. Failed or blocked requests, retries, duplicates and posting URLs that no longer exist are never charged.

You scrape	You pay
100 listings	$0.045
1,000 listings	$0.45
10,000 listings	$4.50
100,000 listings	$45.00

The Apify free plan's $5 monthly credit covers about 11,100 listings. The prefilled input (bike in sfbay, 20 listings) costs $0.009 and took 7 seconds in a cloud test. If you set a maximum cost per run, the scraper stops cleanly when it reaches it.

Why this one

• A quarter of the leaders' price: solidcode $1.80, memo23 $1.50 + $0.007 per run (details and emails extra), fatihtahta $1.49, automation-lab $2.30 + start, ivanvs $6.00 + start, and the $0.50 tier (newbs, unitbytes, ocrad) per 1,000. Here it is $0.45 with details included and no start fee.
• Every Craigslist site: US, Canada, UK, Europe, Asia, or a whole country (US = 413 sites) in one run.
• Past the 10,000 cap: deep search returned 22,027 of 22,027 "furniture" listings in the SF bay area in 29 seconds.
• Your own filters work: paste a craigslist.org search URL and make / model, odometer, bedrooms, condition and the rest are kept.
• Fast and light: the site's own JSON over plain HTTP, up to 360 listings per request, no browser or login.

FAQ

Does it need a Craigslist account or a browser? No. It reads the same public JSON the craigslist.org search page loads (sapi.craigslist.org, rapi.craigslist.org), over HTTPS, without logging in.

How do I find the area name? It's the first part of the address of your local Craigslist: https://seattle.craigslist.org = seattle, https://sfbay.craigslist.org = sfbay. City names like "los angeles" or "SF bay area" work too. Every row has its area and subarea.

How do I use the filters from the Craigslist website? Set the filters on craigslist.org, copy the URL from the address bar and paste it into startUrls. Keyword, area, sub-area, category, price, has-image, posted-today, owner / dealer, postal code + distance, sort and category-specific filters (make / model, model year, odometer, condition, bedrooms, pets, laundry ...) are all kept from the URL.

How do I track new listings for a search? Schedule the actor with sort: "newest" and a maxResults that covers what appears between two runs (for example 50 every hour), and keep the ids you have already seen.

What about personal data? Many posters are private people, so the actor collects no names, phone numbers, e-mail addresses, reply addresses or street addresses. E-mail addresses and phone numbers that posters type into titles and descriptions are replaced with [email removed] / [phone removed]. Map positions are rounded to about 1 km. For job ads, the hiring company name is kept (it is a business).

Is it legal to scrape Craigslist? This actor collects publicly available listing information without logging in, at a polite pace, and stays away from the paths Craigslist's robots.txt disallows (reply, flag, suggest and similar). You're responsible for using the data in line with Craigslist's terms and applicable law. This isn't legal advice; if you're unsure, check with a lawyer.

How does it avoid blocks? It sends normal browser-like requests at a polite pace (about two a second without a proxy), through Apify Proxy when enabled, and rotates sessions when a request is refused (up to 8 retries).

What happens if I run it with no input? It does a demo search for "bike" in the SF bay area (20 listings) instead of failing.

How to use

1. Type one or more search keywords (bike, toyota tacoma, couch) and one or more areas (sfbay, losangeles, chicago, newyork/brk, london). The area is the first part of the site address: sfbay.craigslist.org = sfbay.
2. Optionally add categories (bia bicycles, cta cars & trucks, fua furniture, apa apartments), a price range, or a ZIP code with a distance in miles.
3. Or paste Craigslist URLs: set the filters on craigslist.org, copy the address bar, paste it into startUrls. Category filters (make / model, odometer, bedrooms, condition ...) are kept. Posting URLs work too.
4. Click Start. Listings are saved as they come in; download them as JSON, CSV or Excel, or read them through the Apify API.
5. For a price tracker or new-listing alerts, schedule the actor (for example every hour with sort: "newest" and maxResults: 50) and deduplicate on id.

Input

An empty input runs a 20-listing demo search for "bike" in the SF bay area. Keywords or categories need an area (Craigslist has no nationwide search), a country code, or a postal code.

Field	Default	What it does
searchQueries	-	Keywords as typed into the Craigslist search box, one per line.
areas	-	Craigslist sites: sfbay, newyork, losangeles, seattle, london, toronto ...; sub-areas newyork/brk or brooklyn; US, CA, GB ... for every site in a country (707 sites in total).
categories	sss	Category codes or names: sss all for sale, bia bicycles, cta cars & trucks, cto cars by owner, fua furniture, ela electronics, zip free stuff, hhh housing, apa apartments, roo rooms, jjj jobs, ggg gigs, bbb services ...
postalCode, distanceMiles	-	Search around a ZIP / postal code instead of whole areas (94110, 5 miles). Rows get distanceMiles.
priceMin, priceMax	-	Price range in whole units of the local currency.
hasImage	false	Only listings with photos.
postedToday	false	Only listings posted today.
searchTitlesOnly	false	Keywords must be in the title.
bundleDuplicates	false	Collapse re-posted duplicates into one row (duplicateCount).
sellerType	any	any, owner or dealer (for-sale categories).
sort	newest	newest, oldest, price_asc, price_desc, relevance, distance (with postalCode).
startUrls	-	Craigslist search URLs (all filters kept) and posting URLs (looked up with full details).
maxResults	100	Listings per search (keyword x category x area). A country counts as one area. 0 = all of them.
deepSearch	true	Goes past the 10,000-per-search limit by splitting into sub-areas, categories and price bands.
includeDetails	false	Opens each posting for description, attributes, condition, make / model, year and updated date. Same price, slower.
includeNearbyAreas	false	Also keep the "more from nearby areas" postings Craigslist adds to small searches.

Example: owner-sold bicycles with photos, $100-300, cheapest first, keyword in the title (this exact input returned 100 listings in 1.7 s in our test, all within the price range, all with photos; the site counted 920 matches):

{
    "searchQueries": ["bike"],
    "areas": ["sfbay"],
    "categories": ["bia"],
    "priceMin": 100,
    "priceMax": 300,
    "hasImage": true,
    "searchTitlesOnly": true,
    "sellerType": "owner",
    "sort": "price_asc",
    "maxResults": 100
}

Around a ZIP code, nearest first:

{
    "searchQueries": ["couch"],
    "postalCode": "10001",
    "distanceMiles": 3,
    "sort": "distance",
    "maxResults": 50
}

Several cities at once, or a whole country (maxResults then applies to the country as a whole):

{
    "searchQueries": ["toyota tacoma"],
    "areas": ["losangeles", "sandiego", "phoenix"],
    "categories": ["cta"],
    "maxResults": 200
}

Straight from URLs copied from the browser:

{
    "startUrls": [
        "https://sfbay.craigslist.org/search/sby/cta?auto_make_model=toyota&max_price=15000&sort=priceasc",
        "https://www.craigslist.org/search/area/newyork?cat=apa&min_bedrooms=2",
        "https://sfbay.craigslist.org/pen/bik/d/millbrae-open-wide-gravel-bike-in-xs/7945993826.html"
    ],
    "maxResults": 40
}

Output

One dataset item per listing. This one is from a real run on 19 Oct 2026, search URL https://sfbay.craigslist.org/search/cta?purveyor=owner with includeDetails: true (photo list and attributes shortened):

{
    "id": "7986506352",
    "url": "https://www.craigslist.org/view/d/santa-rosa-porsche-targa-pdk/sh1z1GdmZAgesdP41tDv7e",
    "postingUrl": "https://sfbay.craigslist.org/nby/cto/d/santa-rosa-porsche-targa-pdk/7986506352.html",
    "title": "Porsche 2015 911 4S Targa PDK…",
    "price": 129000,
    "priceText": "$129,000",
    "currency": "USD",
    "category": "cto",
    "categoryName": "cars & trucks - by owner",
    "area": "sfbay",
    "areaName": "SF bay area",
    "subarea": "nby",
    "subareaName": "north bay / marin",
    "location": "Santa Rosa",
    "country": "US",
    "latitude": 38.47,
    "longitude": -122.64,
    "postedAt": "2026-10-20T00:21:08.000Z",
    "imageUrl": "https://images.craigslist.org/00N0N_fuNRyWj0oMP_0CI0t2_600x450.jpg",
    "imageUrls": ["https://images.craigslist.org/00N0N_fuNRyWj0oMP_0CI0t2_1200x900.jpg", "..."],
    "imageCount": 6,
    "odometer": 13600,
    "searchArea": "sfbay",
    "position": 1,
    "scrapedAt": "2026-10-20T00:25:40.517Z",
    "description": "Only 13,480 miles\n\nThe car has Fuchs Rims on it now and -\n\nThe car comes with the -original set of wheels from Porsche stored in there boxes - additional value of $10,000 for the 4 rims.",
    "attributes": { "make/model": "911 4 S Targa  PDK", "year": "2015", "condition": "excellent", "cylinders": "6 cylinders", "fuel": "gas", "odometer": "13,600", "title status": "clean", "transmission": "automatic", "type": "convertible" },
    "condition": "excellent",
    "makeModel": "911 4 S Targa  PDK",
    "year": 2015,
    "titleStatus": "clean",
    "section": "for sale",
    "updatedAt": "2026-10-20T00:21:08.000Z"
}

An apartment and a job from the same run (shortened):

{ "id": "7986506097", "title": "Beyond Four Walls: A Place to Truly Call Home!", "price": 3022, "bedrooms": 2, "sqft": 657, "location": "Emeryville", "subarea": "eby", "housingType": "apartment", "bathrooms": "2", "attributes": { "laundry": "w/d in unit", "parking": "carport", "cats are OK - purrr": true, "dogs are OK - wooof": true } }
{ "id": "7986500661", "title": "HIRING NOW: Office Admin  – (Walnut Creek)", "jobTitle": "Office Admin", "company": "apollo roofing  company", "salary": "TBD", "employmentType": "full-time", "categoryName": "admin/office" }

The dataset has two ready-made views, Overview (with photo thumbnails) and Details. A RUN_SUMMARY record in the key-value store lists, for each search, how many listings were saved, the total the site shows, how many narrower searches deep search ran, why it stopped, and any warnings (for example an area that was not found).

Field fill rates (real test runs)

Search rows (2,500 rows, "bike" in the SF bay area): id, url, postingUrl, title, currency, category, area, subarea, location, latitude / longitude, postedAt, imageCount 100%; price / priceText 96%; photos 95%; neighborhood 71%.

With includeDetails (120 rows: cars by owner, apartments, jobs, furniture): description, attributes, updatedAt 100%; cars: makeModel, year, odometer, titleStatus 100%, condition 73%, VIN 20%; apartments: bedrooms, sqft, bathrooms, housingType 100%; jobs: jobTitle, salary, employmentType 100%, company 63%; furniture: condition 100%, make 57%.

API example

Python (pip install apify-client):

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("rel8ble/craigslist-scraper").call(run_input={
    "searchQueries": ["toyota tacoma"],
    "areas": ["sfbay", "losangeles"],
    "categories": ["cta"],
    "maxResults": 200,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["title"], "|", item["priceText"], "|", item["location"], "|", item["url"])

One HTTP call that runs the actor and returns the rows (curl):

curl -X POST "https://api.apify.com/v2/acts/rel8ble~craigslist-scraper/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"searchQueries": ["iphone 13"], "areas": ["chicago"], "maxResults": 20}'

Integrations

• Make, Zapier and n8n: trigger runs and pass new Craigslist listings into any workflow (Telegram, Slack or e-mail alerts for new cars or apartments under a price).
• Google Sheets: send each run's dataset straight into a spreadsheet to track prices over time.
• Apify API: start runs and fetch results over REST, or with the JavaScript and Python API clients.
• Webhooks: get notified when a run finishes, then pull the new data.
• MCP for AI agents: the Apify MCP server (https://mcp.apify.com) lets Claude, ChatGPT, Cursor and other agents call this actor and read Craigslist listings directly. Typical agent call: {"searchQueries": ["iphone 13"], "areas": ["chicago"], "maxResults": 20}, under 2 cents.
• Schedules: run it every hour with sort: "newest" (or postedToday: true once a day) to catch new listings as they appear.

Limits

• One Craigslist search lists at most 10,000 postings. Deep search (on by default) splits bigger searches by sub-area, then category, then price band, and merges the results without duplicates. A few postings fall outside every narrower slice (for example listings without a price once price bands are needed), so a split search can end slightly below the site's count: everything for sale in the East Bay gave 82,877 of 83,372 (99.4%). The run summary shows the site's total next to what was saved.
• No nationwide search on Craigslist. Use US (or another country code) to search every site in a country; sites are searched in Craigslist's own area order, a few at a time, until maxResults is reached.
• Dates. Search rows carry the posted (or last renewed) time; with includeDetails you also get the time the poster last edited the posting.
• Not included: poster names, e-mail addresses, phone numbers, reply / relay addresses, street addresses, and anything that needs a Craigslist account. The actor never calls Craigslist's reply or contact endpoints.

How it works (for developers)

craigslist.org's search page is a JavaScript app that loads results from sapi.craigslist.org/web/v8/postings/search/full. The first call returns up to 360 postings in a compact array format with lookup tables (decode): posting ids and dates are stored as offsets from minPostingId / minPostedDate, places as indexes into area, place-name and neighborhood tables, and extra fields as tagged arrays ([4, ...photo ids], [6, url slug], [10, price text], [5, bedrooms, sqft], [9, odometer], [7, salary] ...). Asking for the whole list returns up to 10,000 postings without titles plus a cacheId; titles and photos then come from .../search/batch in blocks of 1,080. Posting details come from rapi.craigslist.org/web/v8/postings/<uuid>. Area and category lists are Craigslist's public reference.craigslist.org files.

npm install
npm test                                   # parser tests on saved real API answers
APIFY_LOCAL_STORAGE_DIR=./storage node src/main.js   # input in storage/key_value_stores/default/INPUT.json