Descriptions:
Creator David Ondrej demonstrates how to supercharge Hermes Agent — an open-source AI agent framework with nearly 200,000 GitHub stars — by connecting it to Apify through the Model Context Protocol (MCP). Apify offers more than 40,000 web scraping actors covering platforms like LinkedIn, Google Maps, TikTok, Instagram, and Reddit, most of which block standard browser-based agents. By bridging Hermes and Apify via MCP connectors, the combined setup can retrieve structured data from virtually any website and pipe it directly into downstream workflows.
The tutorial covers the full implementation stack: installing Hermes Agent locally with a single terminal command, selecting an LLM provider (options include OpenRouter, Anthropic, and MiniMax), configuring the Apify actor of choice, generating API keys and MCP connector credentials, and persisting scraped results in a Supabase PostgreSQL table using the SQL editor. The worked example focuses on LinkedIn talent sourcing — using a specific Apify actor that scrapes candidate profiles by job title, location, and industry without requiring LinkedIn credentials or cookies, then deduplicating rows in Supabase by profile URL.
David also shares his personal use case for competitive market intelligence across Europe. The video is sponsored by Apify, and all associated prompts, SQL table presets, and actor configurations are available in a linked resource bundle. Viewers comfortable with a command line but new to MCP or Apify should be able to replicate the full setup within the runtime of the video.
📺 Source: David Ondrej · Published June 15, 2026
🏷️ Format: Tutorial Demo
