# Blog Writing Pipeline

Source: https://openflow.run/en/templates/blog-writing-pipeline/
Content language: en
Template revision: f2bf5895d91486412f320151dafd9c6b95b42091

Marketing

Generate research notes, an outline, a Markdown blog draft, and an edited post with four separate model steps.

[Use in Open Flow](https://console.openflow.run/flows?importUrl=https%3A%2F%2Fraw.githubusercontent.com%2Foomol-lab%2Fawesome-workflow-templates%2Ff2bf5895d91486412f320151dafd9c6b95b42091%2Fworkflows%2Fblog-writing-pipeline%2Ftemplates%2Fblog-writing-pipeline.openflow.json)

*   [Open Flow JSON](https://raw.githubusercontent.com/oomol-lab/awesome-workflow-templates/f2bf5895d91486412f320151dafd9c6b95b42091/workflows/blog-writing-pipeline/templates/blog-writing-pipeline.openflow.json)
*   [n8n JSON](https://raw.githubusercontent.com/oomol-lab/awesome-workflow-templates/f2bf5895d91486412f320151dafd9c6b95b42091/workflows/blog-writing-pipeline/templates/blog-writing-pipeline.n8n.json)

![Blog Writing Pipeline workflow in Open Flow](https://openflow.run/_astro/preview-light.DkhM6guX_2eJgd3.webp)

Produce a Markdown blog post through separate research, outline, drafting, and editing steps. The workflow returns the edited post, topic, and generation time.

## Setup

For Open Flow, select a configured `llama3:8b` model in every model step. For n8n, connect each attached **Ollama Model** to an Ollama credential with the base URL of a service reachable from n8n. Confirm that it serves `llama3:8b`. The native Ollama models send each step’s generation budget as `num_predict`. You can choose another available model in both formats; its wording and quality may differ.

In Open Flow, use Run and supply `topic`, `tone`, and integer `word_count`. For example, use “How our support team handles recurring questions”, “professional but conversational”, and `1500`. In n8n, edit those values in **Set Blog Parameters** and use Execute Workflow.

Four model calls run in order. A failure stops later steps. The workflow does not browse sources, verify citations, or publish the post. Treat research notes as model output and check facts before publication. Word count is a prompt target, not an enforced count.

## Changes to the source

*   Use configured model accounts instead of a machine-local Ollama URL.
*   Fix the source requests that omitted POST, and read message output through the model node contract instead of parsing a nonexistent `data` string.
*   Use real paragraph breaks in prompts and stop if editing returns an empty post.

Adapted from [AI Blog Writer Pipeline with Ollama](https://github.com/enescingoz/awesome-n8n-templates/blob/827110332c21cbe000ab13f6d292d5adb1f1f8db/OpenAI_and_LLMs/AI%20Blog%20Writer%20Pipeline%20with%20Ollama.json) in [enescingoz/awesome-n8n-templates](https://github.com/enescingoz/awesome-n8n-templates).

Source collection notice: © 2024–2026 Enes Cingoz, [CC BY 4.0](https://github.com/enescingoz/awesome-n8n-templates/blob/827110332c21cbe000ab13f6d292d5adb1f1f8db/LICENSE). The source export does not identify a separate template author or license. See [repository licensing](https://github.com/oomol-lab/awesome-workflow-templates/blob/f2bf5895d91486412f320151dafd9c6b95b42091/README.md#licensing) for the scope of OOMOL contributions and third-party material.

## What you’ll need

*   Configured llama3:8b model or compatible replacement
*   n8n Ollama credential
*   Topic, tone, and target word count

*   writing
*   blog
*   ai
*   ollama

[View on GitHub](https://github.com/oomol-lab/awesome-workflow-templates/blob/f2bf5895d91486412f320151dafd9c6b95b42091/workflows/blog-writing-pipeline/README.md)