Development Tools MCP Server

Instructions

Implementation Reference

• src/tools/web-scraping.ts:74-92 (registration)

  Registration of the 'extract_images' tool including its name, description, and input schema in the webScrapingTools array.

  {
    name: 'extract_images',
    description: 'Extract all image URLs from a web page',
    inputSchema: {
      type: 'object',
      properties: {
        url: {
          type: 'string',
          description: 'URL to scrape',
        },
        useBrowser: {
          type: 'boolean',
          description: 'Use browser for dynamic content',
          default: false,
        },
      },
      required: ['url'],
    },
  },

• src/tools/web-scraping.ts:311-318 (handler)

  Handler dispatcher in handleWebScrapingTool for 'extract_images', delegating to dynamic or static scraper based on useBrowser flag.

  case 'extract_images': {
    if (config.useBrowser) {
      const data = await dynamicScraper.scrapeDynamicContent(config);
      return data.images;
    } else {
      return await staticScraper.extractImages(config);
    }
  }

• src/scrapers/static-scraper.ts:131-137 (handler)

  StaticScraper.extractImages handler that invokes scrapeHTML and returns the extracted images.

  /**
   * Extract images from HTML
   */
  async extractImages(config: ScrapingConfig): Promise<string[]> {
    const data = await this.scrapeHTML(config);
    return data.images || [];
  }

• src/scrapers/static-scraper.ts:46-57 (helper)

  Core logic for extracting image URLs using Cheerio within the scrapeHTML method.

  const images: string[] = [];
  $('img[src]').each((_, element) => {
    const src = $(element).attr('src');
    if (src) {
      try {
        const url = new URL(src, config.url);
        images.push(url.href);
      } catch {
        // Invalid URL, skip
      }
    }
  });