Facebook, from the command line

Pulling data out of Facebook usually means a headless browser, a brittle pile of selectors, or the Graph API with its app review and tokens. fb takes a different route: it reads the no-JavaScript HTML surface the way a basic phone browser does, parses it into typed records, and renders them as a table, JSON, CSV, or just URLs.

fb page nasa                       # a Page's full profile
fb page nasa --posts --limit 20    # its twenty most recent posts
fb post <url> --comments           # a post and its comment thread
fb id <anything>                   # classify any Facebook id or URL

The binary is pure Go with no runtime dependencies and no browser. fb reads the public pages Facebook serves to search engines, so there is no login and no cookie. When content is private, fb is explicit about the login wall rather than silently returning nothing.

What you can do with it

• Resolve anything. Turn a slug, numeric id, pfbid token, story_fbid, group id, or a fb.watch/fb.me/share/ short link into a typed identity, then into a full record.
• Stream feeds. Walk a Page, profile, or group feed to any depth, stopping by --limit or by date with --since/--until.
• Read a post in full. Pull a post's text, creation time, reaction, comment, and share counts, its media, and the preview comments Facebook renders.
• Collect media. Stream a handle's photos, videos, and reels, and resolve a single photo or video to its full metadata and playable sources.
• Build datasets. Expand a root into a stream of URLs and crawl them into a local SQLite database you can query with SQL.

Where to go next

• New here? Start with the introduction for the mental model, then the quick start.
• Want to install it? See installation.
• Looking for a specific task? The guides cover how fb reads Facebook, pages and profiles, posts and comments, media, search, and datasets.
• Need every flag? The CLI reference is the full surface.