Bright Data Web MCP

Instructions

Implementation Reference

• browser_tools.js:43-65 (handler)

  The complete handler implementation for the 'scraping_browser_navigate' tool. It defines the tool's metadata, input schema using Zod, and the execute function which obtains a browser page and navigates to the specified URL, returning navigation status, title, and current URL.

  let scraping_browser_navigate = {
      name: 'scraping_browser_navigate',
      description: 'Navigate a scraping browser session to a new URL',
      parameters: z.object({
          url: z.string().describe('The URL to navigate to'),
      }),
      execute: async({url})=>{
          const page = await (await require_browser()).get_page({url});
          try {
              await page.goto(url, {
                  timeout: 120000,
                  waitUntil: 'domcontentloaded',
              });
              return [
                  `Successfully navigated to ${url}`,
                  `Title: ${await page.title()}`,
                  `URL: ${page.url()}`,
              ].join('\n');
          } catch(e){
              throw new UserError(`Error navigating to ${url}: ${e}`);
          }
      },
  };

• browser_tools.js:46-48 (schema)

  Zod schema defining the input parameter 'url' as a required string.

  parameters: z.object({
      url: z.string().describe('The URL to navigate to'),
  }),

• browser_tools.js:307-320 (registration)

  The tool is registered by including it in the exported 'tools' array, which is conditionally populated based on the presence of the API_TOKEN environment variable.

  export const tools = process.env.API_TOKEN ? [
      scraping_browser_navigate,
      scraping_browser_go_back,
      scraping_browser_go_forward,
      scraping_browser_links,
      scraping_browser_click,
      scraping_browser_type,
      scraping_browser_wait_for,
      scraping_browser_screenshot,
      scraping_browser_get_text,
      scraping_browser_get_html,
      scraping_browser_scroll,
      scraping_browser_scroll_to,
  ] : [scraping_browser_activation_instructions];