Skip to content
Having trouble with...
 
Notifications
Clear all

Having trouble with the autoblogger

10 Posts
3 Users
0 Reactions
3 Views
hhajj
(@hhajj)
Posts: 5
Active Member
Topic starter
 

I'm looking to build a workflow that searches specific websites for information on a particular topic, then passes this data to an AI agent. The agent would then use a local Ollama model to generate an article based on the prompt and finally post it to WordPress.

I'm currently facing challenges in extracting the correct information or search results to feed into the AI agent.

Here's the workflow:
{
“name”: “Autoblogger RCF”,
“nodes”: [
{
“parameters”: {
“model”: “qwen2.5:32b-instruct”,
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.lmChatOllama”,
“typeVersion”: 1,
“position”: [
660,
220
],
“id”: “275cb8d5-25ca-44d9-8518-6a1c2a93c78c”,
“name”: “Ollama Chat Model”,
“credentials”: {
“ollamaApi”: {
“id”: “d1gIGjutIE0uyDZl”,
“name”: “Ollama account”
}
}
},
{
“parameters”: {},
“type”: “@n8n/n8n-nodes-langchain.memoryPostgresChat”,
“typeVersion”: 1.3,
“position”: [
840,
220
],
“id”: “67be894c-24d8-4523-ad50-93a2f1d6c643”,
“name”: “Postgres Chat Memory”,
“credentials”: {
“postgres”: {
“id”: “5ceCFsG5OgZTBY5K”,
“name”: “Postgres account”
}
}
},
{
“parameters”: {
“model”: “qwen2.5:32b-instruct”,
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.lmChatOllama”,
“typeVersion”: 1,
“position”: [
1120,
200
],
“id”: “4ef205da-63d8-40b3-a4d3-a4ec229a8386”,
“name”: “Ollama Chat Model1”,
“credentials”: {
“ollamaApi”: {
“id”: “d1gIGjutIE0uyDZl”,
“name”: “Ollama account”
}
}
},
{
“parameters”: {
“promptType”: “define”,
“text”: “=Based on this research content:{{ $(‘Fetch Website Content’).item.json.data }} "Write a comprehensive and engaging article about {{ $(‘When chat message received’).item.json.chatInput }} . The aim of this article is to raise cancer awareness among the general public by providing informative, accessible, and empowering content. Please adhere to the following guidelines to ensure the article is high-quality, professional, SEO-optimized, and not mediocre:nnArticle Content Requirements:nn Research and Accuracy: Utilize up-to-date, evidence-based information from reputable sources such as peer-reviewed journals, major medical institutions (e.g., Mayo Clinic, National Cancer Institute), and recognized cancer organizations (e.g., American Cancer Society, Cancer Research UK). Fact-check all data and statistics meticulously. Include in-text citations and a full list of references at the end. Prioritize reputable, primary sources whenever possible. nn Structure and Formatting:n Title: Craft a clear, concise, and engaging title that accurately reflects the article’s content and incorporates relevant keywords for SEO.n Introduction: Provide a brief overview of the topic, explain its importance and impact on the general public, and clearly state what readers will learn in the article. Aim to capture the reader’s attention immediately.n Body:n Organize the content logically using clear and descriptive subheadings that are also SEO-friendly (incorporating relevant keywords).n Use bullet points or numbered lists to highlight key information, practical tips, and actionable advice.n Where applicable, include practical tips and actionable advice that readers can implement in their daily lives.n Conclusion: Summarize the key takeaways of the article, reinforce the importance of the topic, encourage proactive steps related to cancer awareness or prevention, and include a clear call to action (e.g., ‘Talk to your doctor,’ ‘Learn more at [relevant organization website],’ ‘Share this article’).n References and Additional Resources: List all sources cited within the text in a consistent citation style (e.g., APA, MLA). Provide direct links to further reading, support organizations, and reputable websites where readers can find more information.nnWriting Style and Tone:nn Tone: Maintain a supportive and compassionate tone, acknowledging the emotional impact of cancer and offering hope and encouragement. The overall tone should also be informative and empowering, providing clear, factual information that enables readers to make informed decisions about their health.n Style: Use clear and simple language, avoiding complex medical jargon. If technical terms are necessary, provide clear and concise explanations in plain language. Write in an engaging and accessible conversational style that is easy for a general audience to understand. Aim for a concise article length of approximately 600-800 words, ensuring it is detailed yet not overwhelming.n Voice: Use an active voice and employ the first-person plural (we, our) where appropriate to create a sense of community and shared experience. Be respectful and non-judgmental in your language, avoiding any phrasing that could be perceived as blaming or shaming.nnLanguage:nn Deliver two versions of the article: one written in fluent and accurate English, and a second version translated into fluent and accurate Arabic. Ensure high-quality translation that maintains the original meaning, tone, and structure, using appropriate medical terminology in Arabic.nnCredibility Measures:nn Citations: Cite all sources within the text using a consistent method (e.g., hyperlinking to the source or using bracketed numbers corresponding to the reference list). Provide a full list of references at the end of the article.n Regular Updates: Indicate that this article should be reviewed and updated at least annually to reflect the latest research and guidelines. Include a placeholder for the date of the last update at the beginning or end of the article (e.g., ‘Last Updated: [Date]’).n Author Credentials: Include a placeholder for the author’s name and a link to their LinkedIn profile (e.g., ‘Written by [Author Name], connect on LinkedIn: [LinkedIn Profile URL]’).n Transparency: Include a statement disclosing any potential conflicts of interest, if applicable (e.g., ‘The author declares no conflicts of interest.’). Be transparent about the sources of information and the general review process (e.g., ‘This article was based on research from the sources listed below and reviewed for accuracy.’).nnTechnical Specifications (for web optimization):nn Readability: Use short paragraphs and sentences to enhance readability for online audiences.n Font: Suggest the use of a readable and web-friendly font (e.g., Arial, Helvetica, Open Sans) in a size of 12-14 points.n Multimedia: Suggest relevant and high-quality images, infographics, and videos to enhance understanding and engagement. Ensure all multimedia elements include descriptive alt text for accessibility and SEO.n Links and Navigation: Incorporate relevant internal hyperlinks to other articles within the website (if applicable) and external hyperlinks to reputable sources and related high-quality content. Ensure all links are functional. nn Accessibility: Adhere to web accessibility guidelines (WCAG) to ensure the content is accessible to people with disabilities. Use appropriate headers (H1, H2, H3, etc.), bulleted/numbered lists, and descriptive link texts to aid navigation and screen readers.nnSEO Optimization:nn Conduct keyword research for the chosen topic and strategically incorporate relevant keywords and synonyms throughout the title, headings, subheadings, introduction, body, and conclusion.n Optimize image file names and alt text with relevant keywords.n Ensure the article has a clear and logical structure with appropriate use of headings and subheadings (H1, H2, H3 tags).n Include internal and external links to authoritative sources.n Aim for a readable URL structure.n Consider including a meta description and relevant tags for the article.nnOverall Quality:nn Ensure the final article is free of grammatical errors, typos, and factual inaccuracies.n The article should be well-organized, easy to follow, and provide valuable information to the reader, ultimately contributing to increased cancer awareness.nnPlease provide the two versions of this article (English and Arabic) based on these comprehensive guidelines." ,
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.agent”,
“typeVersion”: 1.7,
“position”: [
820,
-60
],
“id”: “402b6bce-008a-4af5-94cf-0c34b5b49c0a”,
“name”: “English blog writer”
},
{
“parameters”: {
“promptType”: “define”,
“text”: “=Translate this to professional Arabic {{ $json.output }}”,
“options”: {
“systemMessage”: “You are a helpful assistant”
}
},
“type”: “@n8n/n8n-nodes-langchain.agent”,
“typeVersion”: 1.7,
“position”: [
1140,
-20
],
“id”: “f0362b2b-9bcd-471e-9d69-956e5b7f2b0d”,
“name”: “Arabic Blog writer”
},
{
“parameters”: {
“public”: true,
“initialMessages”: “Hi there!

:wave:

My name is RCF. How can I assist you today?”,
“options”: {}
},
“type”: “@n8n/n8n-nodes-langchain.chatTrigger”,
“typeVersion”: 1.1,
“position”: [
-520,
-140
],
“id”: “49f705bb-9375-48a2-9a17-42d0894c0e64”,
“name”: “When chat message received”,
“webhookId”: “aa206c14-a65a-41f1-b87c-8a2dbba7d9b6”
},
{
“parameters”: {
“title”: “={{ $(‘When chat message received’).item.json.chatInput }}”,
“additionalFields”: {
“content”: “={{ $(‘English blog writer’).item.json.output }}”,
“postTemplate”: {
“values”: {}
}
}
},
“type”: “n8n-nodes-base.wordpress”,
“typeVersion”: 1,
“position”: [
1600,
0
],
“id”: “fdd77e84-a95f-4972-a895-9314d0f7689e”,
“name”: “Wordpress”,
“credentials”: {
“wordpressApi”: {
“id”: “D01yLS46YqPxmgIR”,
“name”: “Wordpress account”
}
}
},
{
“parameters”: {
“mode”: “raw”,
“jsonOutput”: “{n "urls": [
{ "url": "https://www.cancer.gov/" },
{ "url": "https://www.cancer.org/" },
{ "url": "https://www.mayoclinic.org/cancer-care" },
{ "url": "https://www.cdc.gov/cancer/index.htm" },
{ "url": "https://www.cancer.net/" },
{ "url": "https://www.nccn.org/" },
{ "url": "https://www.cancerresearchuk.org/" },
{ "url": "https://medlineplus.gov/cancer.html" },
{ "url": "https://www.aacr.org/" },
{ "url": "https://www.breastcancer.org/" }
]
}n”,
“options”: {}
},
“type”: “n8n-nodes-base.set”,
“typeVersion”: 3.4,
“position”: [
-320,
-120
],
“id”: “cd6ac006-9fa6-45c7-9160-a0d0f934e9d0”,
“name”: “Website URL Storage”
},
{
“parameters”: {
“url”: “={{ $json.url }}n”,
“options”: {
“response”: {
“response”: {
“responseFormat”: “text”
}
}
}
},
“type”: “n8n-nodes-base.httpRequest”,
“typeVersion”: 4.2,
“position”: [
260,
-100
],
“id”: “97da7104-58f8-4981-99e9-7db27f741e02”,
“name”: “Fetch Website Content”
},
{
“parameters”: {
“options”: {
“reset”: false
}
},
“type”: “n8n-nodes-base.splitInBatches”,
“typeVersion”: 3,
“position”: [
-40,
-100
],
“id”: “26ced1e8-fafc-4dc9-b353-ca14fec1a18b”,
“name”: “Process Each Website”
},
{
“parameters”: {
“operation”: “extractHtmlContent”,
“extractionValues”: {
“values”: [
{
“cssSelector”: “body”
}
]
},
“options”: {}
},
“type”: “n8n-nodes-base.html”,
“typeVersion”: 1.2,
“position”: [
540,
-100
],
“id”: “97844bc1-fed0-4103-aa48-51406a17badb”,
“name”: “Extract Website Content”
}
],
“pinData”: {},
“connections”: {
“Ollama Chat Model”: {
“ailanguageModel”: [
[
{
“node”: “English blog writer”,
“type”: “ai
languageModel”,
“index”: 0
}
]
]
},
“Postgres Chat Memory”: {
“aimemory”: [
[
{
“node”: “English blog writer”,
“type”: “ai
memory”,
“index”: 0
}
]
]
},
“Ollama Chat Model1”: {
“ailanguageModel”: [
[
{
“node”: “Arabic Blog writer”,
“type”: “ai
languageModel”,
“index”: 0
}
]
]
},
“English blog writer”: {
“main”: [
[
{
“node”: “Arabic Blog writer”,
“type”: “main”,
“index”: 0
}
]
]
},
“When chat message received”: {
“main”: [
[
{
“node”: “Website URL Storage”,
“type”: “main”,
“index”: 0
}
]
]
},
“Arabic Blog writer”: {
“main”: [
[
{
“node”: “Wordpress”,
“type”: “main”,
“index”: 0
}
]
]
},
“Website URL Storage”: {
“main”: [
[
{
“node”: “Process Each Website”,
“type”: “main”,
“index”: 0
}
]
]
},
“Fetch Website Content”: {
“main”: [
[
{
“node”: “Extract Website Content”,
“type”: “main”,
“index”: 0
}
]
]
},
“Process Each Website”: {
“main”: [
[
{
“node”: “Fetch Website Content”,
“type”: “main”,
“index”: 0
}
],

]
},
“Extract Website Content”: {
“main”: [
[
{
“node”: “English blog writer”,
“type”: “main”,
“index”: 0
}
]
]
}
},
“active”: false,
“settings”: {
“executionOrder”: “v1”
},
“versionId”: “a86a5f78-edd2-4fe9-b28e-a6164393a234”,
“meta”: {
“templateCredsSetupCompleted”: true,
“instanceId”: “c7b9543a068b4fd0db62c19ba6292798f3a0ea3bea7a5ce2510c962a66d4ac98”
},
“id”: “AOZqw5IwncW2SbCm”,
“tags”:
}

Information on your callin.io setup

  • callin.io version:
  • Database (default: SQLite):
  • callin.io EXECUTIONS_PROCESS setting (default: own, main):
  • Running callin.io via (Docker, npm, callin.io cloud, desktop app):
  • Operating system:

Please provide the rewritten markdown content *it should be in the markdown format.

 
Posted : 22/03/2025 2:08 am
ThinkBot
(@thinkbot)
Posts: 26
Eminent Member
 

Could you copy your entire workflow (Ctrl+A, Ctrl+C), then click the '</>' button in the forum's reply section and paste the workflow between the backticks that appear?

This would help provide more context regarding what's occurring.

Please share the rewritten markdown content.

 
Posted : 23/03/2025 6:13 pm
hhajj
(@hhajj)
Posts: 5
Active Member
Topic starter
 
 
Posted : 23/03/2025 6:27 pm
ThinkBot
(@thinkbot)
Posts: 26
Eminent Member
 

Could you please rephrase the errors you're encountering or clarify your question? You mentioned you're unable to pass data to the agents. At first glance, the setup appears correct.

Edit: Regarding the loop over items node, if your objective is to write a blog post for each web URL, you should connect the "loop" path to the subsequent nodes, not the "done" path. The "done" path is executed only after all iterations within the loop path have been completed.

 
Posted : 23/03/2025 6:29 pm
hhajj
(@hhajj)
Posts: 5
Active Member
Topic starter
 

I've included the list of sites in the set node, batch splitting "1" to search all the sites for the keyword phrase. Subsequently, all search results must be extracted by the HTML node and sent to the AI agent for article creation.

The workflow receives a phrase, then searches all the websites for information, and finally sends all the gathered information to the ollama model for processing based on the prompt.

 
Posted : 23/03/2025 6:45 pm
ThinkBot
(@thinkbot)
Posts: 26
Eminent Member
 

Could you please rephrase that? I'm not quite sure I understand.

 
Posted : 23/03/2025 7:53 pm
hhajj
(@hhajj)
Posts: 5
Active Member
Topic starter
 

The necessary steps are:
- Receive an input phrase
- Search a list of websites for that specific phrase
- Present the findings
- Provide the results as text to an AI agent
- A local Ollama model will then generate the article using these search results and the prompt
- The article will subsequently be published on a WordPress site

We are a cancer support group aiming to enhance cancer patient education through additional content.
I attempted the workflow but encountered errors that I'm unable to pinpoint. Could you assist?

 
Posted : 24/03/2025 12:39 am
ThinkBot
(@thinkbot)
Posts: 26
Eminent Member
 

I haven't fully fleshed it out, but I believe you would approach it like this.

 
Posted : 24/03/2025 2:22 pm
hhajj
(@hhajj)
Posts: 5
Active Member
Topic starter
 

How can I add 10 websites to callin.io and retrieve research results for all ten?

 
Posted : 24/03/2025 3:01 pm
system
(@system)
Posts: 332
Reputable Member
 

This thread was automatically closed 90 days following the last response. New replies are no longer permitted.

 
Posted : 22/06/2025 3:02 pm
Share: