Atlassian Confluence MCP Server

import { PageSchemaType, PageDetailedSchemaType, } from '../services/vendor.atlassian.pages.types.js'; import { ControllerResponse } from '../types/common.types.js'; import { formatUrl, formatDate, formatHeading, formatBulletList, formatSeparator, formatNumberedList, } from '../utils/formatter.util.js'; /** * Format a list of pages for display * @param pagesData - Raw pages data from the API * @param baseUrl - Base URL for constructing page links * @returns Formatted string with pages information in markdown format */ export function formatPagesList( pagesData: PageSchemaType[], baseUrl: string = '', ): string { if (!pagesData || pagesData.length === 0) { return ( 'No Confluence pages found matching your criteria.' + '\n\n' + formatSeparator() + '\n' + `*Information retrieved at: ${formatDate(new Date())}*` ); } const lines: string[] = [formatHeading('Confluence Pages', 1), '']; // Format each page with its details const formattedList = formatNumberedList(pagesData, (page, _index) => { const itemLines: string[] = []; // Basic information itemLines.push(formatHeading(page.title, 2)); // Create an object with all the properties to display const pageUrl = `${baseUrl}/pages/viewpage.action?pageId=${page.id}`; const properties: Record<string, unknown> = { ID: page.id, Status: page.status, 'Space ID': page.spaceId || 'N/A', Title: page.title, Created: page.createdAt ? formatDate(page.createdAt) : 'N/A', Author: page.authorId || 'Unknown', Version: page.version?.number || 'N/A', URL: formatUrl(pageUrl, page.title), }; // Format as a bullet list itemLines.push(formatBulletList(properties, (key) => key)); return itemLines.join('\n'); }); lines.push(formattedList); // Add standard footer with timestamp lines.push('\n\n' + formatSeparator()); lines.push(`*Information retrieved at: ${formatDate(new Date())}*`); return lines.join('\n'); } /** * Format detailed page information for display * @param pageData - Raw page details from the API * @param markdownBody - Pre-converted markdown content for the page body * @param commentsSummary - Optional comments data to include a summary of recent comments * @returns Formatted string with page details in markdown format */ export function formatPageDetails( pageData: PageDetailedSchemaType, markdownBody: string = '*No content available*', commentsSummary?: ControllerResponse | null, ): string { // Create URL const baseUrl = pageData._links.base || ''; const pageUrl = pageData._links.webui || ''; const fullUrl = pageUrl.startsWith('http') ? pageUrl : `${baseUrl}${pageUrl}`; const lines: string[] = [ formatHeading(`Confluence Page: ${pageData.title}`, 1), '', `> A ${pageData.status} page in space \`${pageData.spaceId}\` created on ${formatDate(pageData.createdAt)}.`, '', formatHeading('Basic Information', 2), ]; // Format basic information as a bullet list const basicProperties: Record<string, unknown> = { ID: pageData.id, Title: pageData.title, 'Space ID': pageData.spaceId, Status: pageData.status, 'Created At': formatDate(pageData.createdAt), 'Author ID': pageData.authorId, 'Parent ID': pageData.parentId || 'None', }; lines.push(formatBulletList(basicProperties, (key) => key)); // Content section lines.push(''); lines.push(formatHeading('Content', 2)); // Use the pre-converted markdown body passed by the controller lines.push(markdownBody); // Comments section if available if (commentsSummary && commentsSummary.content) { lines.push(''); lines.push(formatHeading('Recent Comments', 2)); // Check if there's comment content to display if (commentsSummary.content.includes('No comments found')) { lines.push('*No comments found for this page*'); } else { // Extract a shorter version of the comments that doesn't include the original heading // and footer to integrate better with our page details format let commentsContent = commentsSummary.content; // Remove the original "Page Comments" heading if it exists commentsContent = commentsContent.replace( /# Page Comments\n\n/, '', ); // Remove the footer timestamp if it exists commentsContent = commentsContent.replace( /\n\n---+\n\*Information retrieved at:.*\*/, '', ); // Add the extracted comments content lines.push(commentsContent); // Add a link to view all comments if there are more - check in the content string if ( commentsSummary.content.includes('More results are available') ) { lines.push(''); // Link to view all comments for this page const allCommentsUrl = `${baseUrl}/pages/viewpage.action?pageId=${pageData.id}&showComments=true`; lines.push( `*${formatUrl(allCommentsUrl, 'View all comments on this page')}*`, ); } } } // Labels section lines.push(''); lines.push(formatHeading('Labels', 2)); if (pageData.labels?.results && pageData.labels.results.length > 0) { const labelLines: string[] = []; pageData.labels.results.forEach( (label: { id: string; name: string }) => { labelLines.push(`- **${label.name}** (ID: ${label.id})`); }, ); lines.push(labelLines.join('\n')); } else { lines.push('*No labels assigned to this page*'); } // Links section lines.push(''); lines.push(formatHeading('Links', 2)); lines.push(`- **Web UI**: ${fullUrl}`); lines.push(`- ${formatUrl(fullUrl, 'Open in Confluence')}`); // Add standard footer with timestamp lines.push('\n\n' + formatSeparator()); lines.push(`*Information retrieved at: ${formatDate(new Date())}*`); // Optionally keep the direct link if (fullUrl) { lines.push(`*View this page in Confluence: ${fullUrl}*`); } return lines.join('\n'); }