Turn a chat window into a research helper. Ask a question, and it replies with clear answers, pulling fresh results from the web when needed. Great for marketing teams that need quick topic research, content briefs, and fact checks without opening more tabs.

When a message arrives, the chat trigger sends it to an AI agent. The agent uses an OpenAI chat model with a short memory of the last 20 messages to keep context across follow ups. If live data is required, the agent calls SearchAPI with the query and reads the results. The answer is then written in the same chat, so your team gets one clean response that blends reasoning and current data.

Setup is simple. You only need API keys for OpenAI and SearchAPI and access to n8n. Pick your preferred search engine inside the SearchAPI node and adjust model settings to match your tone. Expect faster turnarounds, like cutting one hour of manual searching down to a few minutes per request. Use it for outlines, competitor snapshots, product FAQ checks, and quick market scans.

• Chat trigger captures each incoming message and starts the flow
• AI Agent decides when to search the web and when to answer from context
• OpenAI chat model composes clear, natural language responses
• Simple Memory stores the last 20 messages to support follow up questions
• SearchAPI tool sends the q parameter from the agent and returns web results
• Swap search engines inside SearchAPI settings without changing the flow
• All parts connect in n8n for a single place to ask and get answers

• Reduce manual web research from 60 minutes to 5 minutes per question
• Automate up to 70 percent of routine fact checks with live search results
• Improve answer freshness by always pulling current web data
• Connect OpenAI and SearchAPI in one chat workspace
• Keep context across 20 recent messages for consistent replies
• Handle up to 5 times more requests per hour by removing manual searching

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenAI and SearchAPI. See the Tools Required section above for links to create accounts with these services.
3. Open the OpenAI Chat Model node. In the Credential to connect with dropdown, click Create new credential. Name it clearly, for example OpenAI Prod. Paste your OpenAI API key from your OpenAI account API page and save.
4. Open the SearchAPI node. In the Credential to connect with dropdown, click Create new credential. Name it clearly, for example SearchAPI Prod. Paste your SearchAPI API key from your SearchAPI account and save.
5. In the OpenAI Chat Model node, confirm the selected model. You can use gpt 4o mini or choose another model that fits your cost and speed needs.
6. In the SearchAPI node, choose your preferred search engine in the node options if available. Leave the q parameter mapping as is so the agent can pass queries automatically.
7. Open the AI Agent node and confirm it is connected to the OpenAI Chat Model, Simple Memory, and SearchAPI tool. This ensures the agent can both remember context and perform web searches.
8. Test in the n8n editor. Click Execute Workflow, open the chat, and ask a question that needs current data. Look at the node logs to confirm the SearchAPI call runs and the final message includes recent information.
9. If searches do not run, check that your SearchAPI key is active and you have credits. If answers miss context, increase the memory window in the Simple Memory node from 20 to a higher number.
10. Activate the workflow for ongoing use. Share the chat entry point with your team so they can ask research questions in one place.