Bright Data Skills

Bright Data Skills is the official agent-skills repository from Bright Data, the web data platform, published under the MIT license. It ships around two dozen ready-to-install skills that teach coding agents how to use Bright Data correctly: scraping pages to clean markdown with bot-detection bypass, structured search results as JSON, building production scrapers, orchestrating the Bright Data MCP server, proxy network code generation, SDK best practices, RAG pipelines, competitive intelligence, price comparison, and SEO audits.\n\nInstallation follows the standard agent-skills flow with npx skills add brightdata/skills, working with Claude Code, Cursor, and Codex. The repo is actively maintained and grew to 265-plus GitHub stars within its first months, reflecting the rush of agent builders wiring real web data into their workflows.\n\nThe honest caveat is commercial: the skills themselves are free and MIT-licensed, but the underlying Bright Data platform is a paid API, so every scrape or search burns account credit. Agents need a Bright Data API token and budget awareness before running the skills at scale.

npx skills add brightdata/skills; requires a Bright Data API token for the underlying platform.