Transform articles into engaging threads

Breaks down blog posts into engaging Twitter threads

What you will receive

Twitter Thread Ready

just now

Your thread is ready to post:

1/ I just published a deep dive on database indexing.

Here's the TL;DR in 5 tweets:

2/ Most developers add indexes after things get slow.

By then, you've already trained users to expect poor performance.

3/ The rule I follow: if you're going to query by it, index it.

Before you write the query, not after.

4/ But here's the trap—over-indexing is just as bad.

Every index slows down writes. Choose wisely.

5/ Read the full post: [link]

---
Copy and post →

How it works

  1. 1Humrun fetches your latest blog post from your RSS feed
  2. 2AI breaks it into a Twitter/X thread format
  3. 3You get numbered tweets ready to post

You configure

https://yourblog.com/feed.xml

The RSS feed URL of your blog

sk-...

For generating thread content

View Python code
import requests
import feedparser
import os

BLOG_RSS_URL = os.environ.get("BLOG_RSS_URL")
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")

# Fetch the latest post
feed = feedparser.parse(BLOG_RSS_URL)

if not feed.entries:
    print("No posts found in feed")
else:
    post = feed.entries[0]
    title = post.get("title", "Untitled")
    content = post.get("summary", post.get("description", ""))
    link = post.get("link", BLOG_RSS_URL)

    # Generate Twitter thread
    prompt = f"""Transform this blog post into a Twitter thread.

Title: {title}
Content: {content[:2000]}
Link: {link}

Guidelines:
- Start with a hook tweet that makes people want to read more
- 5-7 tweets total
- Number each tweet (1/, 2/, etc.)
- Each tweet should be under 280 characters
- Last tweet links to the full post
- No hashtags, no emojis
- Conversational but insightful"""

    response = requests.post(
        "https://api.openai.com/v1/chat/completions",
        headers={"Authorization": f"Bearer {OPENAI_API_KEY}"},
        json={
            "model": "gpt-4o-mini",
            "messages": [{"role": "user", "content": prompt}],
            "max_tokens": 500
        }
    )

    thread = response.json()["choices"][0]["message"]["content"]

    print("Twitter Thread Ready\n")
    print(thread)
    print(f"\n---\nBased on: {title}")
Suggested schedule: Every week on Wednesday•Notifications: After every run
Browse more templates