Skip to content
fb

Quick start

From install to a Page's feed as structured data, in a few commands.

This walks from a fresh install to a Page's feed on your terminal as clean records. It takes a minute.

1. Confirm the binary runs

fb version
fb id nasa -o json

fb id classifies any handle or URL with no network and no login, so it always works and is the fastest sanity check.

{"input":"nasa","kind":"page","page_id":"nasa","slug":"nasa","canonical_url":"https://www.facebook.com/nasa","mbasic_url":"https://mbasic.facebook.com/nasa"}

2. Confirm how fb reads Facebook

fb reads anonymously, as a web crawler, with no login and no cookie:

fb whoami

MODE       USER_AGENT
anonymous  Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)

The read commands below work against public Pages, profiles, groups, and posts. When a target is private, fb exits 4 with a login-wall hint instead of guessing. See how fb reads Facebook for what the crawler surface exposes.

3. Resolve a Page

fb page nasa

NAME  CATEGORY            LIKES   FOLLOWERS  VERIFIED  URL
NASA  Aerospace company   24.1M   25.0M      true      https://www.facebook.com/nasa

4. Stream its feed

fb page nasa --posts --limit 10 -o jsonl

Each line is one full post record. Pipe it into jq, a file, or another tool:

fb page nasa --posts --limit 50 -o jsonl > nasa-posts.jsonl

5. Read a post in full

Take a permalink from the feed and pull its text, counts, and preview comments:

fb post <url> --comments -o jsonl

6. Build a small dataset

Expand the feed into URLs and crawl each into SQLite, then query it:

fb seed page nasa --limit 100 | fb crawl --db nasa.db --comments
fb db --db nasa.db query "select owner_name, count(*) from posts group by 1"

Where to go next

2 min read