article-summarizer

Article Summary

Fetches a web article, generates a concise summary with key takeaways and vault connections, and saves a structured reference note to the Obsidian vault's REFERENCES/ folder.

When to Use

Invoke /article-summarizer <url> when you want to:

• Capture a web article as a reference note in your vault
• Get a concise summary and key takeaways without reading the full piece
• Connect an article's ideas to existing notes in your vault
• Build a searchable library of articles and their learnings

Workflow

1. Discover Vault Root

Read Obsidian configuration to find the active vault:

cat ~/Library/Application\ Support/obsidian/obsidian.json

Parse the JSON to find the vault with "open": true, or use the most recently opened vault. Extract vault_root from the path field.

2. Fetch Article Content and Metadata

Fetch the article in two passes using WebFetch:

Pass A — metadata extraction (structured JSON):

WebFetch(url, prompt="Extract from this article: 1) the article title, 2) author name(s) — omit if not found, 3) publication date in YYYY-MM-DD format — omit if not found, 4) the site or publication name (e.g. 'The Atlantic', 'hbr.org'). Return ONLY a JSON object with keys: title, author, published, source. Use empty string for any field not found.")

Pass B — full article content (clean markdown):

WebFetch(url, prompt="Extract the main article content as clean markdown. Remove navigation menus, ads, sidebars, footers, cookie notices, social sharing buttons, related articles widgets, and comments sections. Preserve the article's headings, body paragraphs, lists, blockquotes, code blocks, and emphasis. Return only the article content.")

Then run the Python script to create the skeleton note:

python3 skills/article-summarizer/fetch_article.py "<vault_root>" "<url>" \
  --title "<title>" \
  --author "<author>" \
  --published "<published>" \
  --source "<source>"

The script outputs JSON to stdout: