📰 XavvyNess Article Intel — Article Extraction + AI Analysis

Extract content from any article URL and get a structured AI intelligence report — summary, key takeaways, sentiment score, actionable insights, topic tags, and follow-up questions. Beats raw article scrapers by adding the analysis layer on top.

6.4 million runs were done with the original Smart Article Extractor. We do everything it does, plus AI analysis — at no extra cost using Groq/Gemini free tier.

---

🚀 What It Does

1. Fetches and cleans the article text from any news, blog, or academic URL
2. Strips ads, nav, footers — extracts only the real content
3. Runs AI analysis tailored to the analysis depth you choose
4. Returns structured JSON ready for downstream pipelines

Perfect for: market research, content monitoring, news aggregation, competitive intelligence, LLM training data, RAG pipelines

---

📥 Input

Field	Required	Default	Description
urls	✅	—	Article URLs to analyze (one or more)
depth	—	standard	quick / standard / deep
includeRawText	—	false	Include full article text in output
language	—	en	Output language for AI analysis

Example input:

{
  "urls": [
    "https://techcrunch.com/2025/01/01/ai-agents-2025/",
    "https://www.wired.com/story/ai-pricing-models/"
  ],
  "depth": "deep",
  "language": "en"
}

---

📤 Output

Real output from a live run on Wikipedia's AI article:

{
  "url": "https://en.wikipedia.org/wiki/Artificial_intelligence",
  "title": "Artificial intelligence - Wikipedia",
  "author": "",
  "publishedAt": "",
  "wordCount": 30139,
  "summary": "Artificial intelligence (AI) refers to the development of computer systems that can perform tasks requiring human intelligence. The field has made significant progress with applications in natural language processing, computer vision, and robotics. AI has the potential to revolutionize many industries.",
  "keyTakeaways": [
    "AI is a field of computer science focused on creating intelligent machines",
    "Machine learning and deep learning are core subfields driving modern AI",
    "Applications span healthcare, finance, transportation, and creative industries",
    "AI raises important ethical questions around bias, privacy, and job displacement",
    "The field has cycles of 'AI winters' and breakthroughs; current era is a breakthrough period"
  ],
  "actionableInsights": [],
  "sentiment": "neutral",
  "sentimentScore": "0.5",
  "topicTags": ["Artificial Intelligence", "Machine Learning", "Computer Science", "Technology", "Innovation"],
  "followUpQuestions": [],
  "audienceType": "",
  "credibilitySignals": [],
  "error": null,
  "analyzedAt": "2026-04-08T22:27:50.945Z",
  "agent": "XavvyNess Article Intel"
}

For news articles with richer content, actionableInsights, followUpQuestions, audienceType, and credibilitySignals are fully populated.

---

💡 Analysis Depths

Depth	Output	Best For
quick	Summary + sentiment + 3 topic tags	High-volume monitoring
standard	+ key takeaways + sentiment score + 5 tags	General research
deep	+ actionable insights + follow-up questions + audience type + credibility signals	Strategic analysis

---

⚙️ Setup — API Keys

Variable	Required	Where to Get
GROQ_API_KEY	Recommended (free)	console.groq.com
GOOGLE_API_KEY	Optional fallback	aistudio.google.com

---

❓ FAQ

Q: What happens if the site blocks crawlers (403)?
A: You'll get a clear error: "Access denied (403) — site blocks automated requests". Try a different article from the same publication, or use a proxy.

Q: Does it handle paywalled content?
A: It extracts whatever is publicly visible. If content is behind a hard paywall, you'll get partial text with a note.

Q: Can I use this for 1000+ articles?
A: Yes — batch your URLs in the urls array. The actor runs up to 3 concurrent fetches by default.

Q: What if AI analysis fails?
A: Falls back automatically: Groq → Gemini → heuristic sentence extraction. Never returns empty-handed.

---

🔗 Use Cases

1. News monitoring — Track mentions of your brand, competitors, or keywords across publications
2. Competitive intelligence — Summarize competitor blog posts and announcements daily
3. Research pipelines — Feed structured article summaries into your RAG or LLM workflows
4. Content strategy — Identify trending topics and sentiment in your space
5. Lead generation — Monitor industry news for buying signals and company announcements
6. SEO research — Analyze top-ranking articles for topic coverage and keyword density

---

📊 Performance

• ✅ Processes most articles in under 10 seconds
• ✅ Handles news sites, blogs, academic papers, and long-form journalism
• ✅ 3-attempt retry with categorized error messages
• ✅ Fallback: Groq → Gemini → heuristic (never silently fails)
• ✅ Failed runs are not charged — you only pay for successfully analyzed articles

---

Integration

Via Apify JavaScript client

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });

const run = await client.actor('4WlWdAOeb33aHxEfY').call({
  urls: [
    'https://techcrunch.com/2026/01/15/ai-agents-enterprise/',
    'https://www.wired.com/story/openai-pricing-2026/',
  ],
  depth: 'standard',
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach(article => {
  console.log(article.title);        // Article title
  console.log(article.summary);     // AI summary
  console.log(article.keyTakeaways); // Bullet-point takeaways
  console.log(article.sentiment);   // positive | neutral | negative
});

Via HTTP API

curl -X POST \
  "https://api.apify.com/v2/acts/4WlWdAOeb33aHxEfY/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "urls": ["https://techcrunch.com/2026/01/15/ai-agents-enterprise/"],
    "depth": "standard"
  }'

Via Make.com / Zapier

Use the Apify module → Run Actor action. Actor ID: 4WlWdAOeb33aHxEfY. Pass your article URLs in the input JSON, then map {{summary}}, {{keyTakeaways}}, and {{sentiment}} from the output to your next step.

---

Built by XavvyNess — AI agent services that do real work.