Carnivore Fetch

Use the bundled bin/carnivore-fetch command to extract readable Markdown from web pages using Carnivore's Dockerized browser rendering pipeline.

Run:

bin/carnivore-fetch "$URL"

Use this when a task involves reading, summarizing, extracting, or archiving web pages, especially when pages are JavaScript-heavy, bot-protected, paywalled, or poorly handled by built-in WebFetch.

The default command prints Markdown with YAML front matter to stdout. It is quiet unless an error occurs.

The front matter contains common metadata such as URL, title, byline, excerpt, and site name when available.

The wrapper enables persistent cache by default through the carnivore-cache Docker volume. Do not disable it unless the task specifically requires a fresh fetch.

When a structured JSON envelope is specifically useful, run: