File Operation MCP Server

import type { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js'; import { z } from 'zod'; import axios from 'axios'; import * as cheerio from 'cheerio'; import { URL } from 'url'; /** * Tool: Scrape Web * Registers the tool with the MCP server * @param server MCP server instance */ const registerTool = (server: McpServer) => { server.registerTool( 'scrape_web', { title: 'Scrape Web', description: 'Scrapes a website for data. Can extract all text, specific elements, links, or element attributes.', inputSchema: { url: z.string().describe("The URL of the website to scrape."), selector: z.string().optional().describe("A CSS selector to target specific elements."), extract_attribute: z.string().optional().describe("The attribute to extract from the selected elements (e.g., 'href', 'src')."), extract_all_text: z.boolean().optional().default(false).describe("If true, extracts all text from the page, ignoring other selectors."), extract_links: z.boolean().optional().default(false).describe("If true, extracts all absolute links from the page."), output_format: z.enum(['text', 'json']).optional().default('text').describe("The desired output format."), } }, async ({ url, selector, extract_attribute, extract_all_text, extract_links, output_format, }) => { try { const { data, request } = await axios.get(url, { responseType: 'text' }); const finalUrl = request.res.responseUrl || url; const $ = cheerio.load(data); let results: string | string[]; if (extract_all_text) { results = $('body').text().replace(/\s\s+/g, ' ').trim(); } else if (extract_links) { results = $('a[href]').map((i, el) => { const href = $(el).attr('href'); if (!href) return null; try { return new URL(href, finalUrl).href; } catch { return null; // Ignore invalid URLs } }).get().filter(Boolean); } else if (selector) { const elements = $(selector); if (extract_attribute) { results = elements.map((i, el) => $(el).attr(extract_attribute)).get().filter(Boolean); } else { results = elements.map((i, el) => $(el).text().trim()).get(); } } else { return { content: [{ type: 'text', text: "Error: You must either provide a 'selector', or set 'extract_all_text' or 'extract_links' to true." }], isError: true, }; } if (output_format === 'json') { return { content: [{ type: 'text', text: JSON.stringify(results, null, 2) }] }; } else { return { content: [{ type: 'text', text: Array.isArray(results) ? results.join('\n') : results }] }; } } catch (error: any) { let errorMessage = `Error scraping web: ${error.message}`; if (error.response) { errorMessage += `\nStatus: ${error.response.status}`; } return { content: [{ type: 'text', text: errorMessage }], isError: true }; } } ); }; export default registerTool;