Use HarvestLab Actors Through Apify MCP

What this setup does

Apify MCP lets compatible clients call HarvestLab actors as tools. Instead of copying data between tabs, the client can prepare input, start an actor run, wait for the dataset, and read the structured output back into the workflow.

Use MCP when you already know which actor matches the job. The public HarvestLab set is strongest for Amazon seller research, marketplace pricing, contact enrichment, YouTube transcripts, local lead lists, Indeed job data, Meta Ads Library research, Product Hunt launches, and Reddit discussion feeds.

Hosted MCP server

Use Apify's hosted MCP server:

https://mcp.apify.com

For a focused toolset, pass one or more HarvestLab actors in the tools query parameter:

https://mcp.apify.com?tools=harvestlab/amazon-scraper,harvestlab/ebay-scraper,harvestlab/contact-extractor

Most teams should connect with Apify OAuth when their MCP client supports it. If the client does not support OAuth, configure the Apify API token through that client's supported MCP header or environment-variable settings. Do not paste private tokens into prompts.

Amazon seller research bundle

https://mcp.apify.com?tools=harvestlab/amazon-scraper,harvestlab/amazon-keyword-rank-tracker,harvestlab/amazon-competitor-price-watchlist,harvestlab/amazon-bestsellers-monitor

Prompt:

Use Apify MCP with HarvestLab Amazon actors. For these ASINs, keywords, and
category URLs, return current product data, keyword rank positions, price or
availability changes, bestseller ranks, source URLs, and the top actions a
seller should review next.

Use this when a seller or agency needs one view across product details, search rank, watchlist changes, and category movement. Keep the first run small: a few ASINs, a few keywords, and one category URL.

Marketplace pricing bundle

https://mcp.apify.com?tools=harvestlab/amazon-scraper,harvestlab/ebay-scraper,harvestlab/bol-com-scraper,harvestlab/amazon-competitor-price-watchlist

Prompt:

Use Apify MCP to compare marketplace prices with HarvestLab actors. Return
product URL, title, price, currency, availability, seller, sold-listing context
when available, and price-change signals for the watchlist items.

Use this for price checks across Amazon, eBay, Bol.com, and watched Amazon ASINs. Separate one-time research from recurring alerts so the output stays readable.

Lead research bundle

https://mcp.apify.com?tools=harvestlab/google-maps-eu-scraper,harvestlab/contact-extractor

Prompt:

Use Apify MCP. Find businesses for this city and category with
harvestlab/google-maps-eu-scraper, enrich businesses that have websites with
harvestlab/contact-extractor, and return company name, website, phone, email,
source URLs, category, and a short qualification note.

Use this for local prospecting and CRM enrichment. Google Maps finds businesses; the Website Contact Extractor checks public websites for richer contact context.

YouTube transcript research bundle

https://mcp.apify.com?tools=harvestlab/youtube-scraper

Prompt:

Use harvestlab/youtube-scraper through Apify MCP. Extract transcripts from these
YouTube URLs, return title, channel, transcript language, transcript text,
timestamped segments, source URLs, and a short topic summary. Keep comments and
AI analysis off unless needed.

Use this when a workflow needs cited video transcripts for search, summaries, research, or support knowledge bases without managing the YouTube API directly.

Paid social research bundle

https://mcp.apify.com?tools=harvestlab/facebook-ads-library-scraper

Prompt:

Use harvestlab/facebook-ads-library-scraper through Apify MCP. Collect Meta Ads
Library results for these advertisers or search terms, then return ad ID, page
name, creative text, spend and impression ranges, countries, active status,
source URLs, and recurring creative angles.

This actor is strongest with an official Meta access token. Treat token setup as configuration, not something to paste into a chat prompt.

Job market intelligence bundle

https://mcp.apify.com?tools=harvestlab/indeed-scraper

Prompt:

Use harvestlab/indeed-scraper through Apify MCP. Search these roles and
locations, return job title, company, location, salary text, remote flag, skills
mentioned, job URL, and a short summary of hiring patterns.

Use this for salary checks, recruiting research, and weekly skills-demand reports. Keep query settings stable when you want trend lines.

Launch and community research bundle

https://mcp.apify.com?tools=harvestlab/producthunt-scraper,harvestlab/reddit-scraper

Prompt:

Use Apify MCP with harvestlab/producthunt-scraper and harvestlab/reddit-scraper.
Find launches and related Reddit discussions for this market, then return launch
URL, makers, topics, upvotes, discussion links, sentiment, and products worth
watching.

Use Product Hunt for launch-side positioning and Reddit for community reaction. Together they are more useful than either source alone.

Practical setup notes

Keep the toolset narrow. A client connected to one or two relevant actors tends to produce cleaner runs than a client connected to a huge generic tool list. Start with one focused MCP URL, inspect the dataset output, then add another actor only when the workflow needs it.

For production work, store run IDs, dataset URLs, and source URLs beside the final records. That makes it possible to audit a summary or report later without rerunning the same scrape.