Extract insights from newsletters you follow

import requests
from bs4 import BeautifulSoup
import feedparser
import os

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

# Try RSS first (most newsletters have one)
rss_urls = [
    NEWSLETTER_URL.rstrip("/") + "/feed",
    NEWSLETTER_URL.rstrip("/") + "/rss",
    NEWSLETTER_URL
]

post = None
for rss_url in rss_urls:
    feed = feedparser.parse(rss_url)
    if feed.entries:
        post = feed.entries[0]
        break

if not post:
    # Fallback: scrape the archive page
    response = requests.get(NEWSLETTER_URL, headers={"User-Agent": "Mozilla/5.0"})
    soup = BeautifulSoup(response.text, "html.parser")

    # Find the first article link
    links = soup.select("article a, .post-preview a, .post a, h2 a")
    if links:
        link = links[0].get("href", "")
        if not link.startswith("http"):
            from urllib.parse import urljoin
            link = urljoin(NEWSLETTER_URL, link)

        article_resp = requests.get(link, headers={"User-Agent": "Mozilla/5.0"})
        article_soup = BeautifulSoup(article_resp.text, "html.parser")

        title_elem = article_soup.select_one("h1, .post-title")
        title = title_elem.get_text(strip=True) if title_elem else "Latest Issue"
        content = article_soup.get_text(strip=True, separator=" ")[:2500]
    else:
        print("Could not find newsletter content")
        exit()
else:
    title = post.get("title", "Latest Issue")
    content = post.get("summary", post.get("description", ""))[:2500]
    link = post.get("link", NEWSLETTER_URL)

# Generate takeaway
prompt = f"""Summarize this newsletter issue into a quick takeaway.

Title: {title}
Content: {content}

Format:
- Key takeaway (1-2 sentences, the main insight)
- Why it matters (1 sentence)

Keep it under 75 words total. Be specific, not generic."""

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": 200
    }
)

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

print(f"Latest: {title}\n")
print(takeaway)
print(f"\nRead more: {link}")