Apify MCP
AI-powered web scraping and automation platform. Run actors, scrape websites, and automate browser tasks via the official Apify MCP server.
Verdict
Common use cases
- Scrape competitor pricing tables weekly
- Monitor job boards for new postings
- Extract contact info from directory sites
- Track product reviews across marketplaces
- Automate form submissions at scale
Integration
- Vendor
- Apify MCP
- Category
- other
- Auth
- OAUTH2
- Composio slug
apify_mcp
Tools
Setup
Setup guide
- 11. In Switchy, open the Space where you want to use Apify and click Integrations in the sidebar. 2. Search for Apify MCP and click Connect. 3. You'll be redirected to Apify's OAuth consent screen — sign in with your Apify account. 4. Grant Switchy permission to run Actors and access datasets on your behalf (read and write scopes). 5. After authorization, you'll return to Switchy and see Apify MCP listed as connected. 6. Test the connection by typing '@Apify list my Actors' in the Space — the AI should return your available scrapers. 7. To run a scraper, @mention Apify with the Actor name and target URL, like '@Apify run Web Scraper on example.com'. 8. Results appear in the thread as JSON or CSV; the AI can parse and summarize them immediately.
What teammates see: by default, memories from Apify MCP are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
List Available Scrapers
@Apify show me all the Actors I have access to, grouped by category if possible.
Scrape Product Listings
@Apify use the Amazon Product Scraper Actor to extract the top 20 results for 'wireless headphones' and return the data as a table.
Monitor Page Changes
@Apify create a monitoring task for https://example.com/pricing that checks daily and notifies me of any changes to the pricing table.
Extract Contact Emails
@Apify run the Email Extractor Actor on https://example.com/team and give me a CSV of all email addresses found.
Summarize Scraped Data
@Apify fetch the last run of my 'Competitor Pricing' dataset and summarize any price changes compared to last week.
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Apify scrape the pricing pages from the top 5 CRM vendors and return a structured comparison of their enterprise tier features
The MCP would return a structured dataset containing pricing information and feature lists extracted from each vendor's website. For example: Salesforce Enterprise at $165/user/month with unlimited custom apps and API calls; HubSpot Enterprise at $1,200/month with advanced reporting and predictive lead scoring; similar entries for three other vendors. Data includes plan names, monthly costs, key features, and source URLs. This illustrative response assumes successful scraping runs; actual output depends on website structure and anti-bot measures at crawl time.
This example demonstrates web scraping at scale using Apify's actor ecosystem. Users need OAuth2 authorization and sufficient Apify platform credits. Scraping jobs run asynchronously — results may take 30 seconds to several minutes depending on target site complexity. Some websites block automated access; the MCP will surface errors when actors fail.
@Apify start a daily monitor on our competitor's blog RSS feed and send me a summary whenever they publish new content about AI features
The MCP would confirm that a scheduled actor has been created with daily execution. Response might include: 'Monitoring actor deployed successfully. Schedule ID: sch_abc123. Next run: tomorrow at 09:00 UTC. The actor will check the RSS feed, filter for posts containing AI-related keywords, and trigger a notification with post titles, publication dates, and excerpt summaries.' This illustrative confirmation shows the setup phase; actual monitoring results arrive as separate notifications over time.
This showcases Apify's scheduling and monitoring capabilities. The MCP creates persistent automation that runs independently of Switchy sessions. Users should understand this incurs ongoing platform costs per execution. OAuth2 scope must include write permissions for actor creation. Monitoring frequency and keyword filters can be adjusted after initial setup.
@Apify extract all product reviews from this Amazon listing URL and analyze the sentiment distribution across 1-star through 5-star ratings
The MCP would execute an Amazon scraper actor and return aggregated review data. Illustrative output: '847 reviews extracted. Sentiment breakdown: 5-star (62%, common themes: fast shipping, good value), 4-star (18%, mixed on durability), 3-star (9%), 2-star (5%), 1-star (6%, frequent complaints about packaging damage). Average rating: 4.3/5. Data includes review text, dates, verified purchase status, and helpful vote counts.' This example shows post-processing; actual response structure depends on the specific Amazon actor used.
This combines data extraction with AI-driven analysis. The MCP handles the scraping; Switchy's AI performs sentiment synthesis from the returned dataset. Amazon's anti-scraping measures mean success rates vary — residential proxies may be required for reliable access. Large product pages with thousands of reviews will hit rate limits and require pagination handling.
Use-case deep-dives
When Apify MCP makes sense for recurring web scraping
A 5-person product team needs to track competitor pricing weekly across 12 SaaS sites that don't offer APIs. Apify MCP is the right call here because it connects pre-built scrapers (called Actors) directly to your Switchy workspace—no custom scraping code, no maintenance when sites change their HTML. The OAuth2 flow means the whole team shares one Apify account without passing API keys around Slack. The threshold: if you're scraping once or twice ever, just use a one-off tool. If you're running the same scrape weekly or daily, and the site is complex (JavaScript-heavy, login-walled), Apify's Actor marketplace saves you 6-10 hours per scraper. Book a call if your team is already paying for Apify and wants the data inside Switchy's context window.
How Apify MCP fits a 3-rep sales team's enrichment workflow
A 3-person sales team is building lists from LinkedIn Sales Navigator and needs email/company data that isn't in their CRM. Apify MCP works if you're already using Apify Actors for enrichment (LinkedIn profile scraper, email finder, company data). The MCP surfaces those scrapers as callable tools in Switchy, so reps can enrich a lead mid-conversation without tab-switching. The trade-off: Apify charges per scrape, so this only pencils if you're enriching 50+ leads a month and the per-lead cost is under your CAC threshold. If you're doing 10 leads a month, manual enrichment is faster. If you're doing 500+, you want a dedicated enrichment API (Apollo, ZoomInfo) with flat-rate pricing, not pay-per-scrape. This is the sweet spot for small teams with spiky, high-intent enrichment needs.
When Apify MCP beats manual doc scraping for support teams
A 4-person support team needs to pull the latest help docs from 8 vendor tools (Stripe, Twilio, AWS) into Switchy's knowledge base every month because those vendors update their APIs faster than the team can track. Apify MCP is the move if those vendors don't offer structured doc exports and you're tired of copy-pasting. You set up an Actor to scrape each vendor's changelog or API reference, then call it from Switchy on a schedule. The boundary: if the vendor has an RSS feed or a proper API for their docs, use that instead—it's free and faster. If they don't, and you're manually refreshing more than twice a month, Apify pays for itself in time saved. This scenario assumes your Apify plan includes enough compute units for 8 monthly scrapes.
Frequently asked
What does the Apify MCP integration do in Switchy?
It connects your Switchy workspace to Apify's web scraping and automation platform, letting your team query scraped data, trigger actors, and pull results directly into AI conversations. Since Apify specializes in extracting structured data from websites at scale, this integration turns that data into a shared resource your team can reference without switching tabs or writing API calls.
Do I need an Apify account to connect this MCP?
Yes. The integration uses OAuth2, so you'll authorize Switchy to access your Apify account during setup. You don't need admin rights to Apify itself, but whoever connects it should have permission to view the actors and datasets your team wants to use. The OAuth flow handles token exchange automatically — no manual API key copying.
Can this MCP trigger new Apify actor runs or only read existing data?
That depends on which tools the MCP exposes once configured. Apify's API supports both triggering runs and fetching results, so the MCP could do either. If your team needs to kick off fresh scrapes from Switchy, check the tool list after connecting — if it's read-only now, that's a design choice by the MCP author, not a platform limit.
How is this different from just using Apify's web app or API directly?
The MCP embeds Apify's capabilities into your AI workspace, so your team can ask questions like "pull the latest product prices from actor X" without leaving Switchy or writing code. The web app is better for configuring actors and monitoring runs; the API is better for custom automation. The MCP is better for ad-hoc data retrieval during team conversations.
Who on my team should connect the Apify MCP?
Whoever owns your Apify account or has visibility into which actors and datasets matter for your work. They'll go through OAuth once, and the whole workspace inherits access. If your team uses Apify for competitive research or lead generation, connect it early so everyone can query that data without asking the ops person to export CSVs.