Confluence MCP

import sanitizeHtml from 'sanitize-html'; /** * Utilities for cleaning and transforming Confluence content */ /** * Extracts plain text from Confluence Storage Format (XHTML) * This is a simple implementation - a production version would use a proper HTML parser */ export function extractTextFromStorage(storageFormat: string): string { if (!storageFormat) return ''; // Simple regex to strip HTML tags return storageFormat .replace(/<[^>]*>/g, ' ') // Replace HTML tags with spaces .replace(/&[a-z]+;/g, ' ') // Replace HTML entities .replace(/\s+/g, ' ') // Normalize whitespace .trim(); } /** * Truncates content to a specified length, preserving word boundaries */ export function truncateContent(content: string, maxLength: number = 8000): string { if (content.length <= maxLength) return content; // Find a good breaking point const breakPoint = content.lastIndexOf(' ', maxLength); if (breakPoint === -1) return content.substring(0, maxLength); return content.substring(0, breakPoint) + '...'; } /** * Optimizes content for AI context windows by removing redundant information * and focusing on the most important parts */ export function optimizeForAI(content: string): string { // This is a placeholder - a real implementation would be more sophisticated return truncateContent(content); } /** * Converts Confluence Storage Format to Markdown * This is a simple implementation - a production version would be more comprehensive */ export function storageFormatToMarkdown(storageFormat: string): string { if (!storageFormat) return ''; // This is a very basic implementation let markdown = storageFormat; // Convert headings markdown = markdown.replace(/<h1[^>]*>(.*?)<\/h1>/g, '# $1'); markdown = markdown.replace(/<h2[^>]*>(.*?)<\/h2>/g, '## $1'); markdown = markdown.replace(/<h3[^>]*>(.*?)<\/h3>/g, '### $1'); markdown = markdown.replace(/<h4[^>]*>(.*?)<\/h4>/g, '#### $1'); markdown = markdown.replace(/<h5[^>]*>(.*?)<\/h5>/g, '##### $1'); markdown = markdown.replace(/<h6[^>]*>(.*?)<\/h6>/g, '###### $1'); // Convert paragraphs markdown = markdown.replace(/<p[^>]*>(.*?)<\/p>/g, '$1\n\n'); // Convert bold markdown = markdown.replace(/<strong[^>]*>(.*?)<\/strong>/g, '**$1**'); markdown = markdown.replace(/<b[^>]*>(.*?)<\/b>/g, '**$1**'); // Convert italic markdown = markdown.replace(/<em[^>]*>(.*?)<\/em>/g, '*$1*'); markdown = markdown.replace(/<i[^>]*>(.*?)<\/i>/g, '*$1*'); // Convert links markdown = markdown.replace(/<a[^>]*href="([^"]*)"[^>]*>(.*?)<\/a>/g, '[$2]($1)'); // Convert lists markdown = markdown.replace(/<ul[^>]*>(.*?)<\/ul>/g, '$1\n'); markdown = markdown.replace(/<ol[^>]*>(.*?)<\/ol>/g, '$1\n'); markdown = markdown.replace(/<li[^>]*>(.*?)<\/li>/g, '- $1\n'); // Remove remaining HTML tags markdown = sanitizeHtml(markdown, { allowedTags: [], allowedAttributes: {} }); // Fix spacing markdown = markdown.replace(/\n\s*\n/g, '\n\n'); return markdown.trim(); }