Key Concepts
- N8N: A drag-and-drop workflow automation platform.
- Scrapeless: An infrastructure company providing tools for AI agents, automation workflows, and web crawling. Offers headless cloud browser, deep SERP API, and universal crawling APIs.
- AI Coder: An AI-powered coding assistant.
- MCP (Multi-Context Processing): A method for integrating external tools and workflows into an AI coder.
- Gemini: A large language model (LLM) used for summarization due to its large context window.
- API Key: A code used to authenticate and authorize access to an API (Application Programming Interface).
- Headless Browser: A browser that runs without a graphical user interface, used for automated web scraping and testing.
- SERP API: An API that provides access to search engine results pages (SERPs).
- Token Context Window: The amount of text an LLM can process at once.
Setting up a Deep Research Workflow with N8N and Scrapeless
1. Introduction
The video demonstrates a workflow that integrates Scrapeless into an N8N workflow to enable an AI coder to perform deep research on any topic. This allows the AI coder to gain context and improve its performance.
2. Scrapeless Overview
- Scrapeless is presented as a key infrastructure provider for AI agents and automation.
- It offers tools like a headless cloud browser, deep SERP API, and universal crawling APIs.
- The presenter emphasizes its efficiency in creating intelligent autonomous systems.
- Actionable Insight: Scrapeless provides the building blocks for creating AI agents and automation platforms.
3. Scrapeless Account Setup
- The video instructs viewers to sign up for Scrapeless using a link in the description to receive free credits and discounts.
- The process involves navigating to the API key management tab in the Scrapeless settings and creating an API key.
- Important: The API key is crucial for authenticating N8N's access to Scrapeless.
4. N8N Workflow Creation
- A new workflow is created in N8N.
- The trigger is set to "Execute by another workflow," enabling integration with an MCP-based workflow.
- An input named "query" is added to the trigger, which will receive the search term from the AI coder.
5. Integrating Scrapeless Search
- The Scrapeless node is added to the N8N workflow.
- The "search" operation is selected.
- The search query is dynamically populated from the "query" input of the trigger node.
- Scrapeless credentials (the API key) are added to the Scrapeless node.
- The "test" option is used to verify the connection and retrieve search results.
- Key Detail: The search results include links, snippets, and other relevant information.
6. Filtering Search Results
- A "split out" node is added to filter the search results and extract only the organic results.
- This step removes unnecessary information and focuses on the most relevant links for scraping.
7. Scraping Web Pages
- Another Scrapeless node is added, this time using the "scrape" operation.
- The link to scrape is dynamically populated from the filtered organic search results.
- The "execute step" option is used to initiate the scraping process.
- Result: The scraped content of the web pages is retrieved.
8. Summarization with Gemini
- An AI agent block is added to summarize the scraped content.
- The Gemini model is selected for summarization due to its large context window and free availability.
- A system prompt is defined to instruct the AI to create a concise summarization based on the original query and the scraped context.
- Example Prompt: "Make me a summarization based on the original query and then summarize the given context based on the question."
9. Integrating with MCP
- A new N8N workflow is created with an MCP trigger.
- A "tool" node is added, selecting the N8N workflow option and connecting it to the deep research workflow created earlier.
- The AI is instructed to enter the input (the search query) into the tool.
- The MCP trigger URL is copied.
10. Connecting to AI Coder (Windsurf, Cursor, Klein)
- The MCP trigger URL is pasted into the AI coder's MCP server configuration.
- The tool is named (e.g., "scrapeless deep research").
- The AI coder is instructed to use the MCP server to research a specific topic (e.g., "how to use cloud code SDK with JavaScript").
- Outcome: The AI coder calls the MCP server, which triggers the N8N workflow, performs the research, and returns a summarized answer.
11. Conclusion
- The video concludes by highlighting the versatility of Scrapeless for building AI agents and automation workflows.
- Viewers are encouraged to explore Scrapeless and build their own superpowered workflows.
- Quote: "You can use it in your daily workflow with coding or if you do something else, you can still use scrapeless to build any kind of AI agent for you or for context and similar tasks and it works amazingly well."
Technical Terms Explained
- N8N: A visual workflow automation tool that allows users to connect different applications and services without coding.
- Scrapeless: A platform providing infrastructure for web scraping and data extraction, essential for AI agents needing real-time information.
- API Key: A unique identifier used to authenticate requests to an API, ensuring secure access to Scrapeless's services.
- Headless Browser: A web browser without a graphical user interface, used for automating web scraping tasks.
- SERP API: An API that provides structured data from search engine results pages, allowing for efficient data extraction.
- Token Context Window: The maximum amount of text that a language model can process at once, crucial for handling large documents or complex queries.
- MCP (Multi-Context Processing): A method for integrating external tools and workflows into an AI coder.
Logical Connections
The video logically connects the following components:
- Problem: AI coders often lack deep context on specific topics.
- Solution: Integrate a deep research workflow using N8N and Scrapeless.
- Implementation: Step-by-step guide on setting up the workflow, including Scrapeless account setup, N8N workflow creation, integration with Gemini for summarization, and connection to an AI coder via MCP.
- Result: The AI coder can now perform deep research and provide more informed and accurate responses.
Synthesis/Conclusion
The video provides a practical guide to enhancing AI coder capabilities by integrating a deep research workflow using N8N and Scrapeless. By automating the process of searching, scraping, and summarizing information, the AI coder can gain valuable context and improve its performance. The use of Gemini for summarization, due to its large context window, is a key element in ensuring comprehensive and accurate results. The integration with MCP allows for seamless incorporation of the workflow into existing AI coder setups. The main takeaway is that by leveraging these tools, developers can create more intelligent and capable AI assistants.
AI summaries can miss context or contain errors. Check important details against the original video.





