Category-Based Business App Recommendation Bot
🎯 Core Function:
Smart App Discovery, Review Summarization, and Organized Telegram Delivery
This workflow creates an intelligent Telegram bot designed to help users find the best business or productivity applications within a specific category (e.g., "CRM," "Project Management," or "Marketing Automation"). It uses BrowserAct to scrape detailed information from a target app marketplace, and Gemini AI to clean, summarize reviews (Pros & Cons), and format the results into rich, image-supported, Telegram-ready posts.
Part 1: BrowserAct Workflow Description
This core module handles the specialized app database search and data extraction:
Targeted Category Search:
The Agent takes two inputs: the desired Category (e.g., "Sales") and Credentials (optional, for accessing gated content).
It navigates to the app marketplace, searches for the user's category, and clicks on the most related category result page.
App Listing Deep Extraction:
The Agent initiates a Loop List to iterate through all app listings visible on the search results page.
For each app, it performs a deep data extraction, capturing essential details:
Name and Company
Full Description
Image/Logo URL
Pros & Cons (if available from the listing/reviews)
Review Link
Data Consolidation:
The raw, structured list of apps and their attributes is bundled into an output array for processing.
Part 2: Automation Integration with Make.com
The Make.com scenario acts as the user interface and content formatter, ensuring scannable, rich-media delivery:
Intent & Search Trigger:
The workflow starts with a Telegram Trigger.
An initial Filter separates general user chat (handled by a simple AI answer) from specific app or product requests.
Notification & Scraping:
An alert is sent to the user ("Searching for apps in X category...").
The BrowserAct node is called to execute the targeted app scraping.
AI Formatting & Media Logic:
The raw app data is sent to a specialized Gemini AI node ("Telegram JSON Engine").
The AI performs critical tasks:
Sanitization: Cleans text, replaces ampersands (&), and ensures HTML is correctly formatted with single quotes.
Media Logic: Analyzes the captured image URL to determine the post type: if it's a valid, non-SVG image, it sets the type to Image; otherwise, it defaults to text and sets media to null.
Content Structuring: Formats the app details into a rich message using Telegram HTML, including Pros & Cons (Verdict section) and a final link.
Iterative Delivery:
The structured JSON array is passed to an Iterator to process each app recommendation individually.
A final Filter checks the type field:
If type: Image, the app is sent using the "Send a Photo" node.
If type: text, the app is sent using the standard "Send a Message" node.
✨ Applicable Scenarios (Use Cases)
Startup & Small Business Advisory: Automatically generate curated lists of recommended software when setting up a new business function (e.g., "Best 5 CRMs for SMBs").
Marketplace Competitor Analysis: Rapidly scrape and summarize all competitors within a specific app category, highlighting their key selling points (Pros) and drawbacks (Cons).
Affiliate Marketing Content: Quickly create structured, high-quality, image-supported review posts for blog content or social media channels.
Internal Tool Discovery: Use the bot within a company to crowdsource or search for solutions to specific internal needs (e.g., "Find apps for better internal communication").
