Get fast answers from the web and simple APIs using two AI agents. One agent scrapes a page to pull fresh content. The other calls a public API to suggest an activity. Great for quick research, support teams, and project planning. A single click runs both paths so you can compare results side by side.

Here is how it works. A manual test trigger sends two prompts into Set nodes. Each Set node feeds an AI Agent powered by OpenAI. The first agent can call a Webscraper Tool that uses Firecrawl to fetch clean page text with only main content. The second agent can call an Activity Tool that hits the Bored API with type and participants as query fields. Results come back to the agent so it can write a clear response without extra glue steps.

You need an OpenAI API key and a Firecrawl API key. The Bored API does not require a key. Expect faster research with less manual copying. Common uses include pulling issue summaries from a page and offering a structured activity idea for a team. Most teams can set this up in minutes and cut manual web checks by more than half.

• Manual test trigger to run both agent paths on demand
• Two AI agents powered by OpenAI Chat Model for natural answers
• Webscraper Tool uses Firecrawl to fetch page content as markdown
• HTTP Activity Tool calls the Bored API with type and participants
• Set nodes define clear prompts for each agent path
• Agent tool descriptions guide when and how each tool is called
• Optimized scraper response with only main content and cleaned tags

• Reduce manual research from 60 minutes to 10 minutes per request
• Streamline web data gathering by up to 70 percent using an agent tool
• Automate 80 percent of basic API lookups for simple questions
• Improve response consistency by pulling only main page content
• Run two use cases in one run to compare answers and save time

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, Firecrawl and Bored API. See the Tools Required section above for links to create accounts with these services.
3. In n8n Cloud, open the OpenAI Chat Model node, choose Credential to connect with, click Create new credential, and paste your OpenAI API key from the OpenAI platform. Save the credential.
4. Open the Webscraper Tool node. In the Credential to connect with dropdown, click Create new credential, choose HTTP Header Auth, and add an Authorization header with your Firecrawl API key. Save the credential.
5. Leave the Webscraper Tool body as is. It already requests only main content and removes media tags for cleaner text.
6. Open the Activity Tool node. Confirm the URL and query parameters type and participants are set. No API key is needed for the Bored API endpoint used.
7. Open the Set nodes and adjust the example prompts if you want different tasks, such as another site to scrape or a different activity type.
8. Click Test workflow in n8n. Watch each agent run. The scraper agent should return page content and a summary. The activity agent should return a suggestion with type and participants.
9. If the scraper call fails with unauthorized, verify the Firecrawl header value and key format. If OpenAI errors with rate limit, try again later or reduce tool calls.
10. Review the agent outputs. If responses are too long or off topic, tighten the Set node prompts or adjust the tool descriptions to guide the agent behavior.