query_wiki
Retrieve Confluence wiki page content using its unique page code.
Instructions
Get Confluence page content by specific code (like F4CjNw).
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | Page code to search for (like F4CjNw) |
Implementation Reference
- src/tools/wiki.ts:22-179 (handler)The main handler function `handleWikiTool` that executes the 'query_wiki' tool logic. It takes a Confluence page code (like F4CjNw), follows redirects to get the page ID, fetches the page content via Confluence API, converts HTML to plain text, and returns the formatted result.
export async function handleWikiTool( name: string, args: Record<string, unknown> ): Promise<Record<string, unknown>> { try { if (name !== 'query_wiki') { throw new Error(`Unknown wiki tool: ${name}`); } const { query } = args as { query: string; }; const confluenceClient = new ConfluenceClient({ host: process.env.JIRA_HOST!, authentication: { basic: { email: process.env.JIRA_EMAIL!, apiToken: process.env.JIRA_API_TOKEN!, }, }, }); // For codes like F4CjNw, follow redirects to get the page ID let pageId = query; // If the query is a code like F4CjNw, follow redirects if (!/^\d+$/.test(query)) { try { // Follow redirects to get the page ID // eslint-disable-next-line no-undef const response = await fetch(`${process.env.JIRA_HOST}/wiki/x/${query}`, { method: 'GET', redirect: 'follow', headers: { 'Authorization': `Basic ${Buffer.from(`${process.env.JIRA_EMAIL}:${process.env.JIRA_API_TOKEN}`).toString('base64')}`, }, }); if (!response.ok) { return { content: [ { type: 'text', text: `Failed to access page with code: ${query}. Status: ${response.status}`, }, ], isError: true, }; } // Extract the page ID from the final URL const finalUrl = response.url; const pageIdMatch = finalUrl.match(/\/pages\/(\d+)/); if (!pageIdMatch) { return { content: [ { type: 'text', text: `Could not extract page ID from URL: ${finalUrl}`, }, ], isError: true, }; } pageId = pageIdMatch[1]; } catch (error) { return { content: [ { type: 'text', text: `Error following redirects: ${error instanceof Error ? error.message : 'Unknown error'}`, }, ], isError: true, }; } } // Get the complete page content using the ID const page = await confluenceClient.content.getContentById({ id: pageId, expand: ['body.storage', 'space', 'version', 'ancestors'], }); // Convert HTML to plain text const plainTextBody = convertHtmlToPlainText(page.body?.storage?.value || ''); return { content: [ { type: 'text', text: `# ${page.title} **ID:** ${page.id} **Space:** ${page.space?.name} (${page.space?.key}) **URL:** ${page._links?.webui} **Author:** ${page.version?.by?.displayName} **Last Modified:** ${page.version?.when} **Code:** ${query} ## Content ${plainTextBody} ## Page Hierarchy ${page.ancestors?.map((ancestor: Models.Content, index: number) => `${' '.repeat(index)}- ${ancestor.title} (${ancestor.type})`).join('\n') || 'No hierarchy available'}`, }, ], }; } catch (error) { let errorMessage = 'Unknown error occurred'; let errorType = 'Unknown'; if (error instanceof Error) { errorMessage = error.message; // Categorize different types of errors for better debugging if (error.message.includes('401') || error.message.includes('Unauthorized')) { errorType = 'Authentication Error'; errorMessage = 'Authentication failed. Please check your JIRA_EMAIL and JIRA_API_TOKEN credentials.'; } else if (error.message.includes('403') || error.message.includes('Forbidden')) { errorType = 'Permission Error'; errorMessage = 'Access denied. You may not have permission to access this Confluence space or content.'; } else if (error.message.includes('404') || error.message.includes('Not Found')) { errorType = 'Not Found Error'; errorMessage = 'Content not found. The page, space, or query may not exist or may not be accessible.'; } else if (error.message.includes('429') || error.message.includes('Rate Limit')) { errorType = 'Rate Limit Error'; errorMessage = 'Too many requests. Please wait before making another request.'; } else if (error.message.includes('500') || error.message.includes('Internal Server Error')) { errorType = 'Server Error'; errorMessage = 'Confluence server error. Please try again later.'; } else if (error.message.includes('CQL') || error.message.includes('query')) { errorType = 'Query Error'; errorMessage = 'Invalid search query. Please check your search parameters.'; } else if (error.message.includes('network') || error.message.includes('timeout')) { errorType = 'Network Error'; errorMessage = 'Network connection failed. Please check your internet connection and try again.'; } else if (error.message.includes('ENOTFOUND') || error.message.includes('ECONNREFUSED')) { errorType = 'Connection Error'; errorMessage = 'Cannot connect to Confluence server. Please check the host URL and network connectivity.'; } } return { content: [ { type: 'text', text: `Error in wiki operation (${errorType}): ${errorMessage}`, }, ], isError: true, }; } } - src/tools/wiki.ts:3-18 (schema)The tool definition including inputSchema for 'query_wiki'. Defines it as a tool with a required 'query' string parameter (page code like F4CjNw).
export function createWikiTools(): Tool[] { return [ { name: 'query_wiki', description: 'Get Confluence page content by specific code (like F4CjNw).', inputSchema: { type: 'object', properties: { query: { type: 'string', description: 'Page code to search for (like F4CjNw)', }, }, required: ['query'], }, }, - src/index.ts:48-60 (registration)Registration of 'query_wiki' tool listing: calls `createWikiTools()` to include the tool in the list of available tools returned by the server.
const wikiTools = createWikiTools(); return { tools: [ ...projectTools, ...issueTools, ...commentTools, ...transitionTools, ...assignmentTools, ...sprintTools, ...wikiTools, ], }; - src/index.ts:107-108 (registration)Routing/dispatch for 'query_wiki' tool calls: when a tool call name starts with 'query_wiki', it delegates to `handleWikiTool()`.
} else if (name.startsWith('query_wiki')) { return await handleWikiTool(name, args || {}); - src/tools/wiki.ts:181-210 (helper)Helper function `convertHtmlToPlainText` that strips HTML tags and converts HTML entities to plain text, used by the handler to process Confluence page content.
function convertHtmlToPlainText(html: string): string { if (!html) return ''; try { // Remove HTML tags and convert HTML entities preserving line breaks let text = html .replace(/<script[^>]*>[\s\S]*?<\/script>/gi, '') // Remove scripts .replace(/<style[^>]*>[\s\S]*?<\/style>/gi, '') // Remove styles .replace(/<br\s*\/?>/gi, '\n') // Convert <br> to line breaks .replace(/<\/p>/gi, '\n\n') // Convert </p> to double line breaks .replace(/<\/div>/gi, '\n') // Convert </div> to line break .replace(/<\/h[1-6]>/gi, '\n\n') // Convert </h1>-</h6> to double line breaks .replace(/<\/li>/gi, '\n') // Convert </li> to line break .replace(/<\/tr>/gi, '\n') // Convert </tr> to line break .replace(/<[^>]*>/g, '') // Remove all other HTML tags .replace(/ /g, ' ') // Convert non-breaking spaces .replace(/&/g, '&') // Convert ampersands .replace(/</g, '<') // Convert less than .replace(/>/g, '>') // Convert greater than .replace(/"/g, '"') // Convert quotes .replace(/'/g, '\'') // Convert apostrophes .replace(/[ \t]+/g, ' ') // Normalize spaces and tabs (but not line breaks) .replace(/\n\s*\n\s*\n/g, '\n\n') // Reduce multiple consecutive line breaks .trim(); return text; } catch (_error) { return html; // If there's an error, return the original HTML } }