Create blog drafts on demand with an AI writer that can also look up fresh sources. It suits marketing teams that need quick, research backed articles without switching tools. The content is returned in a single field that is ready for review or publishing.

An upstream workflow calls this one and passes a topic as query. The Content Creator Agent uses the Anthropic chat model to write the draft. When the model needs facts, it calls the Tavily search tool, which sends a request to the Tavily API and returns up to three results with raw content. The search runs with basic depth and pulls news sources, so the model can quote and summarize current information. The agent blends that research into the copy and outputs text. A Set node then stores the final draft in a response field, while a second Set node returns a clear error message if something fails.

Provide an Anthropic API key and a Tavily API key, and make sure your caller workflow sends the query field. Expect a large cut in first draft time and more consistent tone across posts. Great for content calendars, SEO briefs, and agency production that needs fast turnarounds. Results usually arrive in under a minute and often need only light edits.

• Runs when called by another workflow, so it fits into larger content pipelines
• Anthropic Chat Model writes long form copy from a short topic brief
• Tavily HTTP Request uses POST to the search API with query, basic depth, news topic, raw content, and up to three results
• Agent merges research findings into the draft for better accuracy
• Set node returns a clean response field ready for downstream steps
• Backup Set node outputs a standard error message if writing fails
• Prompt reads the topic from the input field named query

• Reduce first draft time from 2 hours to 10 minutes
• Cut research time by 70 percent with built in web search
• Keep a consistent voice using one system message
• Scale requests by calling the flow from other workflows
• Lower editing changes by 30 percent with structured output

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 Anthropic and Tavily. See the Tools Required section above for links to create accounts with these services.
3. Open the Anthropic Chat Model node. In the Credential to connect with menu, click Create new credential and follow the on screen steps to add your Anthropic API key. Name the credential clearly, then save and select it.
4. Create a Tavily API key in your Tavily dashboard. In the Tavily tool node, replace the value your-api-key in the JSON body with your real key. Keep the endpoint and other fields as provided.
5. Confirm the input mapping. The agent reads the topic from a field named query. Your caller workflow must send an item with a query value.
6. Build a quick test caller workflow: add a Set node with query set to a topic, then add an Execute Workflow node that runs this article generator workflow.
7. Run the test. Open the Response Set node output and check the response field contains the draft. If you see the fallback message, check your Anthropic credential, Tavily key, and that query is present.
8. Tune research settings in the Tavily node if needed. You can change max_results or search_depth to adjust the amount of source material.
9. Adjust the system message in the agent options to match your brand voice. Keep language clear to get steady results.
10. Connect this flow to your main content pipeline. For example, trigger it after a brief is approved, then send the response to your CMS or editor queue.