AI-Powered Autonomous Research Agent using n8n, Gemini & SerpAPI

Unlock In-Depth Automated Research with this AI Agent

This n8n AI Agent transforms your research process by autonomously handling tasks from initial query to final report generation. Simply provide a research topic, and the agent will intelligently:

1. Generate Diverse Search Queries: Using LLMs (like Google Gemini via OpenRouter), it creates multiple targeted search terms to ensure comprehensive coverage.
2. Execute Web & Wikipedia Searches: It queries search engines (via SerpAPI) for real-time web information and taps into Wikipedia for foundational knowledge.
3. Extract Key Information: It processes search results and uses Jina AI to fetch readable content from web pages, then employs LLMs to extract the most relevant context related to your initial query.
4. Synthesize a Report: Finally, it compiles all gathered insights into a well-structured, Markdown-formatted research report, ready for your review.

This empowers solopreneurs, founders, and technical leaders to quickly conduct market research, competitor analysis, technical investigations, or any deep-dive study without manual drudgery.

Key Features & Benefits

• Autonomous End-to-End Research: From chat input to a full report, automate the entire research lifecycle.
• AI-Powered Query Generation: Leverages Google Gemini (via OpenRouter) to brainstorm effective search strategies.
• Multi-Source Data Gathering: Integrates with SerpAPI for current web results and Wikipedia for structured information.
• Intelligent Content Processing: Uses Jina AI for clean webpage data and LLMs for smart information extraction.
• Structured Report Output: Delivers comprehensive findings in an easy-to-read Markdown format.
• Contextual Understanding: Utilizes Langchain memory components to maintain context throughout the research process.
• Extensible: Built on n8n, allowing easy customization and integration with other tools and services.

• An n8n instance (Cloud or self-hosted).
• OpenRouter API Key (e.g., for accessing Google Gemini models).
• SerpAPI API Key.
• Jina AI API Key (used as a Bearer token for HTTP Header Authentication with the r.jina.ai service).

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Configure the 'LLM Response Provider (OpenRouter)' node:
   • Create or select your OpenRouter API credential in n8n.
   • Ensure the selected model (e.g., google/gemini-2.0-flash-001) is appropriate for your research tasks and available on your OpenRouter account.
4. Configure the 'Perform SerpAPI Search Request' HTTP Request node:
   • Create an n8n credential for SerpAPI. In the node, under 'Authentication', select 'Query Auth'. Enter api_key as the 'Name' and your SerpAPI key as the 'Value'. Alternatively, modify the node to use a standard n8n credential if you store the key there and reference it like {{ $credentials.SerpAPI.key }}.
5. Configure the 'Perform Jina AI Analysis Request' HTTP Request node:
   • Create an n8n credential of type 'Header Auth'.
   • Set the 'Name' to Authorization and the 'Value' to Bearer YOUR_JINA_AI_API_KEY (replace YOUR_JINA_AI_API_KEY with your actual Jina AI token).
   • Assign this credential to the 'Perform Jina AI Analysis Request' node.
6. Review and customize the prompts within the Langchain nodes ('Generate Search Queries using LLM', 'Extract Relevant Context via LLM', 'Generate Comprehensive Research Report') to tailor the AI's behavior to your specific needs.
7. The 'Chat Message Trigger' node allows for interactive research initiation. For batch or scheduled research, replace this trigger with a 'Schedule Trigger', 'Webhook Trigger', or another appropriate n8n trigger node.
8. Test the workflow with a sample query to ensure all connections and APIs are working correctly.
9. Activate the workflow.