Skip to main content

​
General

They are identical - bdata is a shorthand alias installed alongside brightdata. Use whichever you prefer.
Yes. The CLI connects to Bright Data’s infrastructure to handle web requests. You can sign up for free and get started with the free tier.
The CLI supports 40+ platforms through the pipelines command, including Amazon, LinkedIn, Instagram, TikTok, YouTube, Facebook, Reddit, Google Maps, Walmart, eBay, and many more. Run brightdata pipelines list to see the full list.
  • Scrape: markdown (default), html, json, screenshot
  • Search: Formatted table (default), json, pretty
  • Pipelines: json (default), csv, ndjson, jsonl
All commands support -o <path> to write output to a file.
Yes. The CLI is fully pipe-friendly. When stdout is not a TTY, colors and spinners are automatically disabled. Use --json for machine-readable output and BRIGHTDATA_API_KEY environment variable for non-interactive authentication.

​
Authentication

Credentials are stored locally on your machine:
OSPath
macOS~/Library/Application Support/brightdata-cli/credentials.json
Linux~/.config/brightdata-cli/credentials.json
Windows%APPDATA%\brightdata-cli\credentials.json
The file is set to mode 0o600 (owner read/write only).
Use the device flow:
brightdata login --device
This prints a URL and a verification code. Open the URL on any device with a browser, enter the code, and authentication completes on the server.
Run brightdata logout followed by brightdata login with the new account. Or pass a different API key directly:
brightdata login --api-key <new_key>

​
Troubleshooting

This means no default zone is configured. Fix it by either:
# Re-run login (auto-creates zones)
brightdata login

# Or set a zone manually
brightdata config set default_zone_unlocker <zone_name>
Your stored API key is no longer valid. Re-authenticate:
brightdata login
Your API key does not have permission for the requested zone or operation. Check zone permissions in the Bright Data control panel.
You’ve hit the rate limit for your zone. Options:
  • Wait a moment and retry
  • Use --async for large jobs to avoid blocking
  • Contact your account manager to increase limits
The default polling timeout is 600 seconds (10 minutes). For large datasets, increase it:
# Per-command
brightdata pipelines amazon_product "<url>" --timeout 1200

# Via environment variable
export BRIGHTDATA_POLLING_TIMEOUT=1200
The CLI auto-detects TTY support. If detection fails, pipe through cat to force plain output:
brightdata scrape https://example.com | cat
Or use --json for clean, parseable output.

​
Configuration

brightdata config set default_format json
Valid values: markdown, json.
Settings are resolved in this order (highest priority first):
  1. CLI flags - e.g., --zone my_zone
  2. Environment variables - e.g., BRIGHTDATA_UNLOCKER_ZONE
  3. config.json - e.g., default_zone_unlocker
  4. Defaults - built-in fallbacks
Delete the configuration directory:
# Linux
rm -rf ~/.config/brightdata-cli/

# macOS
rm -rf ~/Library/Application\ Support/brightdata-cli/
Then run brightdata login to start fresh.