Turn Reddit posts into fact-checked article drafts
An n8n and LangChain workflow ingests Reddit posts, summarizes comment sentiment, drafts articles in different writer styles, and fact-checks with an editor agent. Task: Suggested implementation: Build an n8n workflow to monitor subreddits for 'article' flairs, extract linked text, and summarize comment sentiment. Use AI agents to draft articles in varied styles, apply SEO formatting, and publish to WordPress, while storing embeddings in Supabase. Human role: Review drafts for accuracy and ensure the fact-checking agent does not enter infinite loops.; tools: n8n. Human role: The original post does not say how a person stayed involved.; limitations: Prerequisites include n8n, OpenRouter, WordPress, and Supabase accounts. Known constraints from secondary summaries indicate the workflow may miss image-based news or posts with non-standard flairs. The editor-in-chief fact-checking agent has been reported to loop or fail, requiring human intervention..
Tools
Published on Agent Work Atlas:
Overview
An n8n and LangChain workflow ingests Reddit posts, summarizes comment sentiment, drafts articles in different writer styles, and fact-checks with an editor agent.
Workflow
Suggested implementation: Build an n8n workflow to monitor subreddits for 'article' flairs, extract linked text, and summarize comment sentiment. Use AI agents to draft articles in varied styles, apply SEO formatting, and publish to WordPress, while storing embeddings in Supabase. Human role: Review drafts for accuracy and ensure the fact-checking agent does not enter infinite loops.
Reported result
Actual outcomes are unknown as the source returned a 403 error. Proposed acceptance check: Run a batch of 10 posts and verify that the OpenRouter cost remains under $1 per 100 articles and that the fact-checking agent completes execution without looping.
Human role
The original post does not say how a person stayed involved.
Limitations
Prerequisites include n8n, OpenRouter, WordPress, and Supabase accounts. Known constraints from secondary summaries indicate the workflow may miss image-based news or posts with non-standard flairs. The editor-in-chief fact-checking agent has been reported to loop or fail, requiring human intervention.
Try it
Act as an automation engineer. I need to set up an n8n workflow to convert Reddit posts into WordPress articles. Inputs: [Target Subreddit Name], [WordPress Site URL], [OpenRouter API Key], [Supabase Connection String]. Steps: 1. Create an n8n workflow that polls [Target Subreddit Name] for posts with the 'article' flair. 2. Extract the URL from the post and fetch the article text. 3. Use an AI agent to summarize the Reddit comments for sentiment. 4. Send the article text and sentiment summary to multiple AI writer agents with distinct styles to draft new articles. 5. Apply SEO formatting to the drafts. 6. Publish the final draft to [WordPress Site URL]. 7. Store embeddings in Supabase. Safeguards: - First, run a dry test with 3 posts. - Monitor the 'editor-in-chief' fact-checking agent for infinite loops; if it loops, halt and alert me. - Do not publish automatically; save drafts to WordPress for human review. Output: A JSON log of processed posts, including cost per article and status (success/fail/loop). If inputs are missing, ask for the Subreddit name and API keys before proceeding.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.