Build a Custom Blog Writing Agent With No Code (Langflow)

Download Langflow and open it on your computer. It runs locally, so you can build a custom subagent without setting up a server or writing code just to get started.

In the project area, click New Flow, go to the content generation section, and choose the Blog Writer template.

Pro tip: Templates are the fastest way to learn Langflow.

Change the reference URL first. Use one of your best blog posts so the flow has something real to imitate.

The built-in crawler can also follow more than one page. If you want a broader writing sample, point it at your blog index and increase the crawl depth to 2.

Then add a text input called topic. The reference content shapes the style, and the topic input tells the model what to write next.

Pro tip: Pick a reference post that already sounds how you want future posts to sound. If the source post is weak, the draft will be weak too.

Open the language model settings and click Manage model providers. Add your OpenAI or Anthropic API key, then pick the model you want to run.

In the source test, Anthropic and OpenAI were used. You could also run a model locally with Ollama and use it in Langflow.

Once the reference URL and topic input are wired up, click Playground and test the flow with a topic.

A simple topic prompt is enough for the first run:

how to use AI for marketing

Langflow will pull the reference content, feed the style and topic into the prompt, and return a first draft.

Use the first draft to tighten the flow. If the draft is too generic, add clearer instructions into the prompt template.

No em dashes. Keep it concise. Sound like our site.

Pro tip: Add voice guardrails early. Short lines like the examples above usually do more than a big abstract style paragraph. If you want better output later, add more sources or drop files into a knowledge base. You can also batch multiple article runs once the basic flow is working.

If you want Claude to use this as a subagent, open Share and choose MCP Server.

From there, connect it to Claude. Once it loads, Claude can call the flow as a tool whenever you need a blog draft in that style.

The larger workflow is to use Langflow for the repetitive subagent, then let Claude save its context for the higher-level work.