Templates
Tested programs for common jobs, each in TypeScript, Python and n8n, with what it cost the last time it ran
Each template is a whole program for one job: watch a brand, find creators, read the comments under a post. Download it, set your key, and run it. Every template here ran on the date its card shows, in both TypeScript and Python, and the card says what that run cost.
export SOCIALCRAWL_API_KEY=YOUR_API_KEY
npx tsx brand-listening.ts # TypeScript, Node 22 or later
python3 brand-listening.py # Python 3.9 or later, nothing to installThe n8n file is a workflow. Import it in n8n, then create a Header Auth credential named SocialCrawl API key, with Name x-api-key and your key as Value, and pick it on every request node.
The top of each program says what a run should cost. ITEMS caps how many rows it collects, and the price scales with it. When it ends, it prints what it spent. An agent can fetch the same programs for free from GET /v1/utility/recipes?id=brand-listening.
Brand and topic listening
Watch what the web says about a brand
For: Brand mentions
For one brand, one window: 50 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 50 credits in 1 call (the quote held 50). The TypeScript and Python programs both ran.
This week's posts about a brand on the platforms you name
For: Reddit and X chatter about a brand this week
For one page from each of two platforms: 2 credits held up front (settles 0-2; the unused hold is refunded).
Measured on 25/10/2026: 2 credits in 1 call for ITEMS=25 (the quote held 2). The TypeScript and Python programs both ran.
Who mentions or replies to an account
For: Who is talking about @handle
For first page of each default search: 3 credits held up front (settles 1-3; the unused hold is refunded).
Measured on 25/10/2026: 1 credit in 1 call for ITEMS=3 (the quote held 3). The TypeScript and Python programs both ran.
Share of voice between competing brands
For: Share of voice for us vs competitors
For three brands with the social leg: 120 credits held up front (settles 20-120; the unused hold is refunded).
Measured on 25/10/2026: 120 credits in 1 call for ITEMS=3 (the quote held 120). The TypeScript and Python programs both ran.
Search a topic on several platforms at once
For: Find TikTok, Instagram and YouTube posts about a topic
For one page from each of three platforms: 3 credits held up front (settles 0-3; the unused hold is refunded).
Measured on 25/10/2026: 3 credits in 1 call for ITEMS=25 (the quote held 3). The TypeScript and Python programs both ran.
Collect thousands of tweets for a keyword
For: Cost of 10k tweets for a keyword
For ten thousand tweets: 1 credit per page (page size not yet measured; quote the whole job with GET /v1/utility/estimate or scripts/estimate.py).
Measured on 25/10/2026: 17 credits in 1 call for ITEMS=25 (the quote held 1). The TypeScript and Python programs both ran.
Find posts from people ready to buy
For: Find people looking for a product like ours
For one page of posts: 1 credit held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 1 credit in 1 call for ITEMS=25 (the quote held 1). The TypeScript and Python programs both ran.
Forum threads about a topic, with top replies
For: What do Reddit and Hacker News say about X
For one fused page of threads: 10 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 10 credits in 1 call for ITEMS=25 (the quote held 10). The TypeScript and Python programs both ran.
News coverage of a topic
For: Latest news about a company
For one call at the default depth: 1 credit held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 1 credit in 1 call for ITEMS=10 (the quote held 1). The TypeScript and Python programs both ran.
Discover
How people reacted to a brand and its competitors last month
For: A monthly report on how people react to a brand and its competitors
For a brand and its competitors, one month: at most 150 credits (max_credits caps the whole run; each search bills at its platform's price only when it returns rows, so it settles 0-150; nothing beyond the judging bound is held up front).
Measured on 18/11/2026: 40 credits in 4 calls (the quote held 150). The TypeScript and Python programs both ran.
Find the accounts behind a topic: sellers, creators, pages and channels
For: Find accounts selling a product or a service on social media
For the accounts behind one topic: at most 400 credits (max_credits caps the whole run; each search bills at its platform's price only when it returns rows, so it settles 0-400; nothing beyond the judging bound is held up front).
Measured on 25/10/2026: 24 credits in 1 call (the quote held 400). The TypeScript and Python programs both ran.
Creators
Instagram creators from one country on a small budget
For: Cheapest way to find German Instagram creators about matcha
For one page of reels with each creator checked: 31 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 26 credits in 26 calls for ITEMS=25 (the quote held 26). The TypeScript and Python programs both ran.
Vet a creator before a partnership
For: Is this influencer legit
For one creator: 50 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 50 credits in 1 call (the quote held 50). The TypeScript and Python programs both ran.
Find a creator's email
For: Find this creator's email
For one creator: 25 credits held up front (exact; the unused hold is refunded).
Measured on 02/11/2026: 0 credits in 1 call (the quote held 25). The TypeScript and Python programs both ran.
One creator on every platform
For: Find this creator's TikTok, Instagram, YouTube and X
For one handle on four platforms: 5 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 5 credits in 1 call for ITEMS=4 (the quote held 5). The TypeScript and Python programs both ran.
A creator's profile, recent posts and the comments on them
For: Get a creator's profile, latest videos and their comments
For profile, one page of videos, first comment page on five videos: 7 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 7 credits in 7 calls for ITEMS=5 (the quote held 7). The TypeScript and Python programs both ran.
Find the social accounts of a person or company
For: What are this company's social accounts
For one name: 17 credits held up front (settles 2-17; the unused hold is refunded).
Measured on 25/10/2026: 17 credits in 1 call (the quote held 17). The TypeScript and Python programs both ran.
Profiles
A LinkedIn member's profile and recent posts
For: LinkedIn profile and posts for a person
For one member, default post window: 10 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 10 credits in 2 calls for ITEMS=20 (the quote held 10). The TypeScript and Python programs both ran.
Find LinkedIn members by role and company
For: Find marketing directors at a company on LinkedIn
For one page of ten members, three backgrounds: 40 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 40 credits in 4 calls for ITEMS=3 (the quote held 40). The TypeScript and Python programs both ran.
Look up a list of handles in one call
For: Vet a spreadsheet of creator handles
For fifty handles: about 50 credits held up front (settles 1-50; the unused hold is refunded); a batch is priced at a cycled example body, so quote your real list first.
Measured on 25/10/2026: 2 credits in 1 call for ITEMS=25 (the quote held 25). The TypeScript and Python programs both ran.
Comments
Export the latest N comments on a TikTok video
For: Get/export/download comments on this TikTok
For two hundred comments: ≥ 15 credits held up front (settles 5-15; the unused hold is refunded); a count or fan-out is unknown, so this is a floor: quote the whole job with GET /v1/utility/estimate or scripts/estimate.py.
Measured on 25/10/2026: 3 credits in 1 call for ITEMS=25 (the quote held 3). The TypeScript and Python programs both ran.
Every comment on any post, in one call
For: All comments on this YouTube video
For five hundred top-level comments: 10 credits held up front (settles 2-10; the unused hold is refunded).
Measured on 25/10/2026: 5 credits in 1 call for ITEMS=25 (the quote held 10). The TypeScript and Python programs both ran.
Which comments are questions, complaints or buyers
For: What are people saying under this post
For one page of comments: 1 credit held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 1 credit in 1 call for ITEMS=25 (the quote held 1). The TypeScript and Python programs both ran.
Transcribe a video
For: Transcript of this YouTube video
For one video: 3 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 3 credits in 1 call (the quote held 3). The TypeScript and Python programs both ran.
Commerce and reviews
Compare one product's reviews across retailers
For: Amazon vs Walmart reviews for one product
For first review page at two retailers: 16 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 16 credits in 4 calls for ITEMS=2 (the quote held 16). The TypeScript and Python programs both ran.
A Home Depot product's price and reviews
For: Price and reviews for a drill at Home Depot
For one search page and the first review page: 10 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 10 credits in 2 calls for ITEMS=2 (the quote held 10). The TypeScript and Python programs both ran.
One report on a product's reviews across marketplaces
For: Pros and cons of a product from its reviews
For one product: 30 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 30 credits in 1 call (the quote held 30). The TypeScript and Python programs both ran.
Reviews of a mobile app
For: App Store reviews for an app
For one app, first review page: 10 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 10 credits in 2 calls (the quote held 10). The TypeScript and Python programs both ran.
Reviews of a local business
For: Yelp reviews for a restaurant
For one search page, one business's first review page: 6 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 6 credits in 2 calls (the quote held 6). The TypeScript and Python programs both ran.
See the ads a competitor is running
For: What ads is a competitor running
For one page of ads: 5 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 5 credits in 1 call (the quote held 5). The TypeScript and Python programs both ran.
Monitoring
Track a creator's new posts every day
For: Poll a creator's new posts daily
Creating is free. Each scheduled run bills 2 credits (instagram/profile/posts read plus 1 scheduling credit; estimated_cost_per_run on create); 1 run per day ≈ 2 credits/day. Running the read by hand once holds 1 credit.
Measured on 25/10/2026: 1 credit in 2 calls (the quote held 1). The TypeScript and Python programs both ran.
Weekly alert on a competitor's new YouTube videos
For: Weekly YouTube competitor monitor
Creating is free. Each scheduled run bills 2 credits (youtube/channel/videos read plus 1 scheduling credit; estimated_cost_per_run on create); 1 run per week ≈ 2 credits per week. Running the read by hand once holds 1 credit.
Measured on 25/10/2026: 1 credit in 2 calls (the quote held 1). The TypeScript and Python programs both ran.
Alert when a web page changes
For: Alert me when a pricing page changes
Creating and reading are free. Each check bills at least 2 credits (the page read, at least 1 credit, plus 1 for orchestration; the page read can bill more); 96 checks per day at least 192 credits/day.
Measured on 25/10/2026: 0 credits in 2 calls for ITEMS=25 (the quote held 0). The TypeScript and Python programs both ran.
Lists of URLs
Engagement stats for a list of post URLs
For: Views, likes and comments for 500 URLs
For five hundred URLs in chunks of a hundred: about 500 credits held up front (settles 5-500; the unused hold is refunded); a batch is priced at a cycled example body, so quote your real list first.
Measured on 25/10/2026: 2 credits in 1 call for ITEMS=25 (the quote held 25). The TypeScript and Python programs both ran.
Cohorts
Which of my accounts posted about a keyword
For: Which of our customers' accounts posted about our brand
For two hundred accounts, two pages each: a cohort query holds its full computed ceiling (estimated_credits), which depends on the members, platforms and page budget Creating the cohort, uploading members, polling and reading results are free.
Measured on 25/10/2026: 4 credits in 24 calls for ITEMS=25. The TypeScript and Python programs both ran.
Jobs and finance
Pay for a role in a country, and the open jobs
For: What do senior data engineers earn in Germany
For one salary lookup and one page of listings: 11 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 11 credits in 2 calls for ITEMS=2 (the quote held 11). The TypeScript and Python programs both ran.
A stock's current price and latest headlines
For: Current price of NVIDIA stock and its news
For one instrument: 7 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 7 credits in 3 calls (the quote held 7). The TypeScript and Python programs both ran.
Web pages
Crawl a site and get each page as markdown
For: Crawl the first 50 pages of a docs site
For fifty pages: 50 credits held up front (settles 1-50; the unused hold is refunded).
Measured on 25/10/2026: 50 credits in 9 calls for ITEMS=25 (the quote held 50). The TypeScript and Python programs both ran.
Pull structured fields from a web page
For: Extract the prices from this page
For one page: 5 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 5 credits in 1 call (the quote held 5). The TypeScript and Python programs both ran.
Korea
What Korean consumers say about a product, and search interest
For: What are Korean consumers saying about a product
For one brief and one trend series: 15 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 15 credits in 2 calls for ITEMS=2 (the quote held 15). The TypeScript and Python programs both ran.
What Korean people ask and say about a topic: questions, communities, blogs, search interest and video
For: What Korean patients or consumers ask about a topic
For one full page from each of five sources: 9 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 9 credits in 5 calls for ITEMS=5 (the quote held 9). The TypeScript and Python programs both ran.
Read what Korean people write on a named community (theqoo, pann, clien, dcinside and others)
For: Korean community posts about a topic on theqoo, pann, instiz, clien or dcinside
For ten posts from one community: 11 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 10 credits in 11 calls for ITEMS=10 (the quote held 11). The TypeScript and Python programs both ran.
Public prices for a cosmetic or clinic treatment in Korea
For: How much does a treatment cost at Korean clinics
For one event list and ten platform pages: 12 credits held up front (exact; the unused hold is refunded).
Measured on 25/10/2026: 12 credits in 12 calls for ITEMS=10 (the quote held 12). The TypeScript and Python programs both ran.
