webscraping_ai_text

Implementation Reference

• src/index.js:285-309 (handler)

  The tool handler function for 'webscraping_ai_text'. It calls client.text() with the URL and options, then returns the response via createSanitizedResponse.

  server.tool(
    'webscraping_ai_text',
    {
      url: z.string().describe('URL of the target page.'),
      text_format: z.enum(['plain', 'xml', 'json']).optional().default('json').describe('Format of the text response.'),
      return_links: z.boolean().optional().describe('Return links from the page body text.'),
      ...commonOptionsSchema
    },
    async ({ url, text_format, return_links, ...options }) => {
      try {
        const result = await client.text(url, {
          ...options,
          text_format,
          return_links
        });
  
        const content = typeof result === 'object' ? JSON.stringify(result) : result;
  
        return createSanitizedResponse(content, url);
      } catch (error) {
        const errorObj = JSON.parse(error.message);
        return createSanitizedResponse(JSON.stringify(errorObj), url, true);
      }
    }
  );

• src/index.js:287-292 (schema)

  Input schema for the 'webscraping_ai_text' tool, defining url, text_format, return_links, and common options.

  {
    url: z.string().describe('URL of the target page.'),
    text_format: z.enum(['plain', 'xml', 'json']).optional().default('json').describe('Format of the text response.'),
    return_links: z.boolean().optional().describe('Return links from the page body text.'),
    ...commonOptionsSchema
  },

• src/index.js:285-309 (registration)

  Registration of the 'webscraping_ai_text' tool using server.tool() with the MCP SDK.

  server.tool(
    'webscraping_ai_text',
    {
      url: z.string().describe('URL of the target page.'),
      text_format: z.enum(['plain', 'xml', 'json']).optional().default('json').describe('Format of the text response.'),
      return_links: z.boolean().optional().describe('Return links from the page body text.'),
      ...commonOptionsSchema
    },
    async ({ url, text_format, return_links, ...options }) => {
      try {
        const result = await client.text(url, {
          ...options,
          text_format,
          return_links
        });
  
        const content = typeof result === 'object' ? JSON.stringify(result) : result;
  
        return createSanitizedResponse(content, url);
      } catch (error) {
        const errorObj = JSON.parse(error.message);
        return createSanitizedResponse(JSON.stringify(errorObj), url, true);
      }
    }
  );

• src/index.js:97-101 (helper)

  The client.text() helper method that makes the actual HTTP request to the /text endpoint of the WebScraping.AI API.

  async text(url, options = {}) {
    return this.request('/text', {
      url,
      ...options
    });

• src/index.js:192-207 (helper)

  The createSanitizedResponse helper used by the handler to format and optionally sandbox the response content.

  function createSanitizedResponse(content, url, isError = false) {
    if (isError) {
      return {
        content: [{ type: 'text', text: content }],
        isError: true
      };
    }
  
    // Process the content (apply sandboxing if enabled)
    const result = sanitizer.sanitize(content, { url });
  
    // Create response
    return {
      content: [{ type: 'text', text: result.content }]
    };
  }