Bright Data CLI
Search and page collection share an agent workflow, with sample data visible before proposed scraper fixes are applied.
- 6,215 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Browser commands click controls and fill forms in a live session; `--auto-approve` applies a generated scraper repair without manual review.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/bright-data-cliDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Browser commands click controls and fill forms in a live session; `--auto-approve` applies a generated scraper repair without manual review.
This is third-party code. Review the repository files before installing.
What it does
Bright Data CLI searches the web, retrieves page content and runs prebuilt scrapers. An agent selects a command for search results, page text or structured records. Separate commands control a browser session, including clicks and form entry. Custom scrapers can be generated or repaired from a description. The user first identifies incorrect data and describes the required correction. After `brightdata scraper heal`, the service returns a sample result and normally waits for approval before applying the fix.
Who it is for. Analysts of authorized web sources collecting data with an agent and evaluating scraper output before further use.
Good fit when
- Research starts from search results for an approved topic.
- An analyst exports selected pages as structured records for comparison.
- An existing scraper fits the source and required format.
- After a page change, a user compares a proposed collector fix with expected data.
Not a fit when
- The workflow must stay entirely local without using Bright Data services.
- The selected data and account are not authorized for use.
- You expect guaranteed results for every website and page state.
Example request
Collect approved pages with an existing scraper, show sample output and stop before modifying the collector or submitting a browser form.Limitations
The CLI requires Node.js 20 or later and a Bright Data account with access to the selected services. `brightdata add mcp` uses the key saved by `brightdata login`; setting `BRIGHTDATA_API_KEY` alone is insufficient for that command. The `--auto-approve` flag applies a scraper repair without human preview. Data availability and structure depend on the site and scraper; results are not guaranteed for every page.
How to disable. Close the browser session with `brightdata browser close`, stop CLI jobs and run `brightdata logout` to remove saved credentials. Zones and scrapers remain in the account.
FAQ
What should I do with awaiting_approval?
Inspect `preview_result` and compare the sample with the expected data. The command in `next_step` applies the fix; `brightdata scraper approve <collector_id> --reject` rejects it.
Does the CLI decide a scraper is broken?
No. The user evaluates output and describes the defect.
Where do I inspect account usage?
`brightdata budget` shows balance and usage information.
How do I end a browser session?
Use `brightdata browser close`.
Does first login change the account?
Yes. The CLI creates `cli_unlocker` and `cli_browser` service zones if absent. `brightdata zones` lists active zones.
Discussion
No questions or reviews yet.
Related
CLI toolsOfficial ScrapeAtlas terminal client: an agent or script pulls public data from 49 platforms with one command and gets plain JSON
Busabase CLI backs up a shared workspace; restoring the archive requires an empty destination prepared by the user.
A CLI and 13 skills for Claude Code: the Power BI data model, DAX measures, row-level security and PBIR report pages straight from chat
A local MCP server that turns a folder of documents into a knowledge base with hybrid search for Claude Code and Cursor
SearchCLI is a Volcengine AI Search tool that evaluates results against test queries and retains an experiment plan and report.
A Claude Code skill that picks a Scrapling Fetcher per site, handles cookie logins and Cloudflare, and writes scraping scripts.
README in short
The CLI installs agent skills and MCP connections for Claude Code, Cursor and Codex. A local process keeps browser sessions alive between commands.