A chat assistant helps marketing teams research topics and draft copy faster. You ask questions in a simple chat, and it pulls facts from the web to support clear answers. It works well for campaign planning, content briefs, and quick fact checks.

A manual chat trigger opens the conversation inside n8n. An AI Agent powered by OpenAI runs the dialogue, using SerpAPI for live search and Wikipedia for background facts. A memory buffer saves the last 20 messages, so the bot remembers what you asked earlier. The model uses a low temperature of 0.3 to keep tone steady and reduce random output.

Setup needs API keys for OpenAI and SerpAPI. After connecting the tools, start a chat and ask for a short brief or a list of sources to confirm it is working. Teams often cut research time from an hour to a few minutes and get consistent notes for blogs, ads, and email copy.

• Manual Chat Trigger starts real time conversations inside n8n
• OpenAI chat model gpt 4o mini with temperature set to 0.3 for steady tone
• AI Agent routes tasks and decides when to search or read Wikipedia
• SerpAPI tool fetches live search results and recent pages
• Wikipedia tool retrieves quick background facts
• Window Buffer Memory keeps the last 20 messages for context

• Reduce research time from 60 minutes to 5 minutes per topic
• Automate about 70 percent of early fact finding
• Improve answer consistency with a 0.3 temperature setting
• Keep context across the last 20 messages to avoid repeats
• Pull facts from web search and Wikipedia in one place

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, SerpAPI and Wikipedia. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, click New and create an OpenAI API Key credential. In your OpenAI account, generate an API key and paste it into the credential. Give it a clear name like OpenAI Main.
4. Open the Chat OpenAI node and select your OpenAI credential. Set the model to gpt-4o-mini and temperature to 0.3.
5. In the n8n credentials manager, create a SerpAPI credential. In your SerpAPI dashboard, copy your API key and paste it into the credential. Name it SerpAPI Prod.
6. Open the SerpAPI node and choose the SerpAPI credential. Set any search parameters you prefer, such as location or language.
7. Wikipedia usually needs no credential. Leave it as is and make sure it is connected to the AI Agent as a tool.
8. Open the Window Buffer Memory node and confirm the context window is set to 20. Ensure it connects to the AI Agent as AI memory.
9. Open the AI Agent node. Confirm prompt type is Define and the Text field reads from {{$json.input}}. Verify links to Chat OpenAI, SerpAPI, and Wikipedia under the agent tool and model sections.
10. Start a chat using the Manual Chat panel. Ask a question that needs fresh data, such as Find recent email open rate benchmarks.
11. Check that the reply references web results or Wikipedia facts. If it does not, ensure both tool nodes are connected to the AI Agent as AI tools.
12. If you see errors or empty replies, verify API keys, usage quotas, and network access. Review the n8n execution log for details and try a simpler query to isolate issues.