Advanced Multi-Source AI Research with Bright Data, OpenAI, and Redis

This n8n workflow automates an advanced AI research process, leveraging external data sources, large language models, and data storage for comprehensive analysis. It's designed to fetch data, process it with an AI agent, store intermediate results, and notify users of the outcome.

What it does

This workflow orchestrates a sophisticated AI research pipeline through the following steps:

1. Receives a Trigger: The workflow is initiated by an external webhook call, likely containing parameters for the research task.
2. Prepares Data: It uses a "Data table" node, which might serve as a placeholder or a static dataset for initial processing.
3. Executes AI Agent: An "AI Agent" node (powered by Langchain) takes the prepared data and processes it using a "Basic LLM Chain" and an "OpenAI Chat Model". This agent is likely responsible for understanding the research query, breaking it down, and utilizing various tools to gather information.
4. Interacts with External Tools: The "AI Agent" can interact with an "MCP Client Tool", which suggests integration with a Model Context Protocol for potentially fetching or pushing data to external services (like Bright Data for web scraping, as hinted by the directory name).
5. Parses AI Output: A "Structured Output Parser" processes the results from the AI agent, ensuring the output is in a consistent, usable format.
6. Stores Intermediate Results: The workflow stores data in a "Redis" database, likely for caching, session management, or persistent storage of research findings.
7. Aggregates Data: An "Aggregate" node combines data, possibly from multiple sources or iterative AI steps, into a single collection.
8. Splits Data: A "Split Out" node then separates the aggregated data into individual items for further processing or distribution.
9. Conditional Logic: An "If" node introduces conditional branching, allowing the workflow to take different paths based on the research results or specific criteria.
10. Sends Notifications (Success): If the condition is met (e.g., research successful), an email is sent via the "Send Email" node, and a message is posted to "Slack".
11. Sends Notifications (Failure/Alternative): If the condition is not met, a different "HTTP Request" is made, and a message is posted to "Slack", indicating an alternative outcome or potential issue.
12. Responds to Webhook: Finally, the workflow sends a response back to the initiating webhook, confirming completion and potentially including results.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• OpenAI API Key: For the "OpenAI Chat Model" to function.
• Redis Instance: Access to a Redis database for data storage.
• SMTP Credentials: For the "Send Email" node to send notifications.
• Slack Account & API Token: For the "Slack" node to post messages.
• Model Context Protocol (MCP) Client: Configuration for the "MCP Client Tool" to interact with external services (e.g., Bright Data for data collection, as suggested by the directory name).
• Webhook Trigger: An external system capable of sending HTTP POST requests to trigger the workflow.

Setup/Usage

1. Import the Workflow: Download the provided JSON and import it into your n8n instance.
2. Configure Credentials:
   • Set up your OpenAI API Key credential for the "OpenAI Chat Model".
   • Configure your Redis credentials for the "Redis" node.
   • Add your SMTP credentials for the "Send Email" node.
   • Set up your Slack API token or webhook for the "Slack" node.
   • Configure the "MCP Client Tool" with the necessary credentials or API endpoints for your Model Context Protocol integration.
3. Configure Webhook:
   • Activate the "Webhook" trigger node and copy its URL.
   • Configure your external system to send POST requests to this URL to initiate the workflow.
4. Customize Nodes:
   • Adjust the "Data table" node if you need to provide specific initial data or modify its structure.
   • Fine-tune the "AI Agent", "Basic LLM Chain", and "Structured Output Parser" nodes with prompts, models, and schema definitions relevant to your research tasks.
   • Modify the "Redis" node operations (e.g., SET, GET, HSET) as needed for your data storage strategy.
   • Update the "If" node conditions to reflect your desired branching logic.
   • Customize the content and recipients for the "Send Email" and "Slack" notification nodes.
   • Adjust the "HTTP Request" node for any alternative actions or error reporting.
   • Ensure the "Respond to Webhook" node returns the desired output to the calling system.
5. Activate the Workflow: Once configured, activate the workflow to start automating your AI research.