ChatGPT, Gemini, Copilot & AI Mode answers

Wulfcare Data scrapers: Google Trends, YouTube, App Store, Google Play, Steam, Shopify, Google Jobs, Hotels, Flights and AI search APIs

Ready-to-run scrapers for data that has no good official API, with Python examples, a pytrends replacement and one-line setup for AI agents (MCP). Each one runs on Apify, so you don't manage proxies, IP blocks or rate limits: you send JSON, you get JSON.

AI agents: start with AGENTS.md (which scraper for which request, one curl call each). Machine-readable: catalog.json, llms.txt, llms-full.txt, full input reference per scraper in reference/, OpenAPI spec per scraper at https://api.apify.com/v2/acts/wulfcare~<name>/builds/default/openapi.json.

Common problems these fix:

• pytrends 429 Too Many Requests / TooManyRequestsError → google-trends (drop-in TrendReq, plus more than 5 keywords on one scale and daily data for any range)
• youtube-transcript-api RequestBlocked / IpBlocked: YouTube is blocking requests from your IP → pip install youtube-transcript-unblocked and change one import (package), or call the Actor directly: youtube-transcripts
• Paying $0.01–$0.05 per transcript on another Apify YouTube transcript Actor → same input and output for $0.003 (change only the Actor id)
• YouTube Data API quotaExceeded (about 100 searches a day) → youtube
• YouTube comments and replies with no API key or quota → youtube-comments
• Paying more on the most used Apify YouTube comments Actor → same input and output field names for $0.25 per 1,000 (change only the Actor id)
• App Store reviews stuck at 500 per app (RSS feed cap) → app-store
• Google Play Developer API only returns your own apps' reviews from the last week → google-play
• No official Google Jobs API → google-jobs
• No official Google Hotels API → google-hotels
• No official Google Flights API → google-flights (prices, price tracking, total trip cost)
• No API for what ChatGPT search, Gemini, Copilot and Google AI Mode actually tell users about your brand → ai-search
• Every Steam review of a game, any language, with no API key → wulfcare/steam-reviews-scraper
• No official Shopify App Store reviews API → wulfcare/shopify-app-store-reviews-scraper
• Flight price drop alerts on a schedule (drops, rises, all-time low, under-target) → wulfcare/flight-price-tracker
• YouTube Shorts views and likes from searches or channels → wulfcare/youtube-shorts-scraper
• Everything trending on Google right now, any country → wulfcare/google-trends-trending-now
• AI training job listings spread over 15 sites → ai-training-jobs (free)

Guides: youtube-transcript-api IpBlocked · pytrends 429 · YouTube quotaExceeded · export YouTube comments to CSV · YouTube comments scrapers compared

Free history: PastCharts saves App Store and Google Play top charts and Google trending searches every day, free to browse or download as JSON and CSV. AI agents can query it with no key through its MCP server at https://pastcharts.com/mcp. Use the scrapers here when you need today's data.

All scrapers

Failed requests and empty results are never charged. Apify's free plan includes $5 of credit every month.

YouTube transcripts in Python (easiest)

1. Get a free Apify token: sign up at https://console.apify.com/sign-up, then Settings → API & Integrations → copy the token ($5 free credit every month).
2. Install and set the token:

pip install youtube-transcript-unblockedexport APIFY_TOKEN=apify_api_...        # Windows PowerShell: $env:APIFY_TOKEN="apify_api_..."

3. Get a transcript:

from youtube_transcript_unblocked import YouTubeTranscriptApi
transcript = YouTubeTranscriptApi().fetch("https://www.youtube.com/watch?v=UF8uR6Z6KLc")
print(" ".join(s.text for s in transcript))   # the whole transcript as textfor s in transcript:                           # or line by line with timestamps    print(f"{s.start:7.1f}s  {s.text}")

Other languages: fetch(video, languages=["es", "en"]) (Spanish, else English). Many videos in one go: fetch_many([...]). Already using youtube-transcript-api? Only the import changes; your existing code keeps working, including on AWS/GCP/Azure where YouTube blocks it. Cost: $0.003 per transcript, videos without captions free. Details: packages/youtube-transcript-unblocked.

Quick start (Python, any scraper)

pip install apify-clientexport APIFY_TOKEN=your_token   # free account: https://console.apify.com/sign-up → Settings → API & Integrations

from wulfcare import run   # wulfcare.py in this repo
rows = run("wulfcare/google-trends-scraper", {"searchTerms": ["coffee, tea"], "geo": "US"})

Every folder has a working example. The full input of each Actor is on its Apify page (Input tab).

Any language (HTTP)

curl -X POST "https://api.apify.com/v2/acts/wulfcare~google-trends-scraper/run-sync-get-dataset-items?token=$APIFY_TOKEN" \  -H "Content-Type: application/json" \  -d '{"searchTerms": ["coffee, tea"], "geo": "US"}'

Swap the Actor name and the JSON input for any scraper above. For runs longer than 5 minutes, start the run with POST /v2/acts/{actor}/runs and read the dataset when it finishes (API docs).

Use them from AI agents (MCP)

Give Claude, Cursor, VS Code or any MCP client these scrapers as tools with Apify's MCP server.

One-click bundles (also listed in the official MCP Registry as io.github.wulfcare/<name>). Paste one URL into your MCP client; each scraper in it becomes one tool:

All of them at once, or any mix you choose:

Hosted (sign in with OAuth, or send Authorization: Bearer <APIFY_TOKEN>):

https://mcp.apify.com?tools=wulfcare/google-trends-scraper,wulfcare/youtube-transcript-scraper,wulfcare/youtube-scraper,wulfcare/youtube-comments-scraper,wulfcare/google-jobs-scraper,wulfcare/google-hotels-scraper,wulfcare/google-flights-scraper,wulfcare/app-store-scraper,wulfcare/google-play-scraper,wulfcare/chatgpt-search-scraper,wulfcare/gemini-search-scraper,wulfcare/google-ai-mode-scraper

Local (claude_desktop_config.json, .cursor/mcp.json, ...):

{  "mcpServers": {    "wulfcare": {      "command": "npx",      "args": ["-y", "@apify/actors-mcp-server", "--tools",               "wulfcare/google-trends-scraper,wulfcare/youtube-transcript-scraper,wulfcare/youtube-scraper,wulfcare/youtube-comments-scraper,wulfcare/google-jobs-scraper,wulfcare/google-hotels-scraper,wulfcare/google-flights-scraper,wulfcare/app-store-scraper,wulfcare/google-play-scraper,wulfcare/chatgpt-search-scraper,wulfcare/gemini-search-scraper,wulfcare/google-ai-mode-scraper"],      "env": { "APIFY_TOKEN": "your_token" }    }  }}

Keep only the ones you need; each becomes one tool. Agents can also read llms.txt for a plain-text summary of every scraper.

Notes

• Everything is collected from public pages as a logged-out visitor sees them. No accounts or logins are used.
• Results are what each source showed at run time and can contain errors; check anything important before you rely on it.
• Questions or a field you need: open an issue here or on the Actor's Issues tab on Apify.

MIT licensed (the examples and helpers in this repo).