LLMs need text, not hero images

HTML embeds are for humans; **plain content** is for chunking, embeddings, and summarization. One call returns body text without nav, clap bars, or script tags.

1 min read No scrapers required Updated Jul 9, 2026

Pipeline

Discover ids via user feed or search
Fetch /article/{id}/content
Chunk → embed → store in vector DB
Optionally fetch info for title/tags metadata

Compliance

Respect Medium terms and author rights. Many teams only index their own posts or licensed partners.

Endpoints

/article/{article_id}/content
/article/{article_id} (metadata)

Starter code

javascript

const articleId = 'f5ef1da2850d';
const res = await fetch(`https://api.zenndra.com/article/${articleId}/content`, {
  headers: { Authorization: `Bearer ${process.env.ZENNDRA_API_KEY}` },
});
const { content } = await res.json();
// chunk `content` for your vector store