bright-data-mcp — Claude Code skill: web data via MCP
This skill routes every web data task in Claude through the Bright Data MCP server instead of the built-in WebFetch and WebSearch. Useful when pages block bots or you want clean JSON.
npx claude-code-templates@latest --skill web-data/bright-data-mcpWhat bright-data-mcp does
The skill tells the agent which Bright Data MCP tool to call for a given job: search, page reading, structured data from supported platforms, browser automation. It ships a decision tree and a full tool reference in references/mcp-tools.md, plus connection steps in references/mcp-setup.md.
Two modes. Rapid is the default and covers search_engine, scrape_as_markdown and their batch variants. Pro unlocks more tools, among them web_data_* for Amazon, LinkedIn, Instagram, TikTok and YouTube, and the scraping_browser_* set; it needs the pro=1 parameter on the remote MCP URL.
Check that the MCP server is connected before the first call — in Claude it has to be configured in your settings. No connection, no skill: the agent falls back to built-in tools. One thing to weigh: the skill explicitly forbids falling back to WebFetch or WebSearch, so on a team that mandates built-in tools only, that's worth a look before you install.
What it helps with
- pull data from a site that blocks ordinary requests
- get structured JSON instead of parsing markdown by hand
- run up to ten search queries in one call
- collect content from several URLs in a batch
- click and type through a page when plain reading is not enough
Example prompts for the agent
- Use Bright Data MCP to collect reviews for this Amazon product and return a table of ratings
- Search the web for recent articles about this framework with search_engine and save the links to a file
- Open the product page in the Bright Data browser, click the size selector and grab the price
How to install bright-data-mcp
1claude-code-templates
npx claude-code-templates@latest --skill web-data/bright-data-mcp2Skills CLI (Vercel)
npx skills add davila7/claude-code-templates --skill bright-data-mcp -a claude-code3Manually: put the files in place
~/.claude/skills/bright-data-mcp/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/bright-data-mcp/
Manually: put the files in place
.agents/skills/bright-data-mcp/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/bright-data-mcp/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install davila7/claude-code-templates bright-data-mcp2Manually: put the files in place
.github/skills/bright-data-mcp/SKILL.mdPlatform CLI
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/web-data/bright-data-mcpManually: put the files in place
.devin/skills/bright-data-mcp/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about bright-data-mcp
How do I install bright-data-mcp in Claude?
Through the templates CLI: npx claude-code-templates@latest --skill web-data/bright-data-mcp. Or with the skills CLI: npx skills add davila7/claude-code-templates --skill bright-data-mcp -a claude-code. The skill file then lives at ~/.claude/skills/bright-data-mcp/SKILL.md.
What does bright-data-mcp do?
It sets a rule that all web data work goes through the Bright Data MCP server: search, page scraping, structured data from supported platforms, browser automation. Built-in WebFetch and WebSearch are left unused.
What do I need before using it?
A connected Bright Data MCP server in your Claude settings. Pro mode additionally requires the pro=1 parameter on the remote MCP URL.