JIRA MCP Server

import { Version3Client } from "jira.js"; import dotenv from "dotenv"; dotenv.config(); const JIRA_URL = process.env.JIRA_URL; const JIRA_EMAIL = process.env.JIRA_EMAIL; const JIRA_API_TOKEN = process.env.JIRA_API_TOKEN; if (!JIRA_URL || !JIRA_EMAIL || !JIRA_API_TOKEN) { console.error("Error: Missing required environment variables:"); console.error("- JIRA_URL (e.g., https://your-domain.atlassian.net)"); console.error("- JIRA_EMAIL"); console.error("- JIRA_API_TOKEN"); process.exit(1); } const client = new Version3Client({ host: JIRA_URL, authentication: { basic: { email: JIRA_EMAIL, apiToken: JIRA_API_TOKEN, }, }, }); /** * Extract text content from JIRA document structure * @param {Object} node - Document node (paragraph, text, etc.) * @returns {string} - Extracted text content */ function extractTextFromNode(node) { if (!node) return ""; if (node.type === "text") { let text = node.text || ""; // Check for hyperlink marks if (node.marks && Array.isArray(node.marks)) { const linkMark = node.marks.find((mark) => mark.type === "link"); if (linkMark && linkMark.attrs && linkMark.attrs.href) { // Include hyperlink URL in the text const url = linkMark.attrs.href; // Format as "text [URL]" if text exists, otherwise just show URL return text ? `${text} [${url}]` : `[${url}]`; } } return text; } if (node.type === "hardBreak") { return "\n"; } if (node.type === "paragraph" && node.content) { return node.content.map(extractTextFromNode).join(""); } // Handle bullet lists if (node.type === "bulletList" && node.content) { return node.content.map((item) => extractTextFromNode(item)).join("\n"); } // Handle ordered lists if (node.type === "orderedList" && node.content) { return node.content.map((item) => extractTextFromNode(item)).join("\n"); } // Handle list items - prefix with bullet or number if (node.type === "listItem" && node.content) { const itemText = node.content.map(extractTextFromNode).join(""); return `• ${itemText}`; } if (node.content && Array.isArray(node.content)) { return node.content.map(extractTextFromNode).join(""); } return ""; } /** * Extract text from comment body (document structure) * @param {Object} body - Comment body object * @returns {string} - Plain text content */ function extractCommentText(body) { if (!body || !body.content) return ""; return body.content .map((node) => extractTextFromNode(node)) .join("\n") .trim(); } /** * Filter function to transform JIRA API response to only include specific fields * @param {Object} jiraResponse - The raw JIRA API response object * @returns {Object} - Filtered response with only required fields */ function filterJiraResponse(jiraResponse) { if (!jiraResponse || !jiraResponse.issues) { return []; } const filteredIssues = jiraResponse.issues.filter((issue) => { // 1. Check last comment const comments = (issue.fields?.comment?.comments || []).sort((a, b) => new Date(b.updated) - new Date(a.updated)); const lastCommentBody = comments.length > 0 ? extractCommentText(comments[0].body) : ""; if (lastCommentBody.toLowerCase().includes("completed")) { return false; } // 2. Check linked issues const issuelinks = issue.fields?.issuelinks || []; const allLinkedIssuesReady = issuelinks.every((link) => { // User requested to only check inward linked issues (issues that link TO this issue) if (!link.inwardIssue) return true; const inwardIssue = link.inwardIssue; const inwardIssueType = inwardIssue.fields?.issuetype?.name; // If the inward linked issue is a Bug or LO Bug, check its status if (inwardIssueType === "Bug" || inwardIssueType === "LO Bug") { const status = inwardIssue.fields?.status?.name; return status === "Ready for QA" || status === "Closed"; } // If it's not a Bug/LO Bug, we don't care about its status return true; }); if (!allLinkedIssuesReady) { return false; } return true; }); // 3. Format return return filteredIssues.map((issue, index) => { const summary = issue.fields?.summary || ""; // Construct URL from self link (e.g. https://scopely.atlassian.net/rest/api/3/issue/2151081 -> https://scopely.atlassian.net/browse/ST-156821) // A safer way might be to just use the base domain if we can infer it, or just use the key if the user knows the domain. // However, the user asked for "JIRA Ticket URL". // Let's try to parse the domain from 'self' or just assume a standard structure if 'self' is present. let ticketUrl = ""; if (issue.self) { try { const url = new URL(issue.self); ticketUrl = `${url.origin}/browse/${issue.key}`; } catch (e) { ticketUrl = issue.key; // Fallback } } else { ticketUrl = issue.key; } return `${index + 1}. ${ticketUrl} - ${summary}`; }); } /** * Test function to fetch real JIRA data and test the comment extractor */ async function testCommentExtractor() { try { console.log("=== Fetching JIRA issues ==="); console.log('JQL: project = ST AND issuetype = Task AND component in (BuildReq, ValidationReq) AND status = "In Progress" AND updated >= -90m\n'); const response = await client.issueSearch.searchForIssuesUsingJqlEnhancedSearch({ jql: 'project = ST AND issuetype = Task AND component in (BuildReq, ValidationReq) AND status = "In Progress" AND updated >= -90m', fields: ["summary", "status", "comment", "updated", "key", "issuelinks"], maxResults: 50, startAt: 0, }); console.log(`Found ${response.issues?.length || 0} issues\n`); // Test comment extraction on each issue if (response.issues && response.issues.length > 0) { response.issues.forEach((issue, index) => { console.log(`\n${"=".repeat(80)}`); console.log(`Issue ${index + 1}: ${issue.key} - ${issue.fields?.summary}`); console.log(`${"=".repeat(80)}`); const comments = issue.fields?.comment?.comments || []; if (comments.length > 0) { // Sort by updated date to get the latest comment const sortedComments = comments.sort((a, b) => new Date(b.updated) - new Date(a.updated)); console.log(`\nTotal comments: ${comments.length}`); console.log(`\n--- Latest Comment (by ${sortedComments[0].author?.displayName}) ---`); const extractedText = extractCommentText(sortedComments[0].body); console.log(extractedText); // Check if it contains "completed" if (extractedText.toLowerCase().includes("completed")) { console.log("\n⚠️ This comment contains 'completed' - would be filtered out"); } } else { console.log("\nNo comments found"); } }); // Run the filter function console.log("\n\n" + "=".repeat(80)); console.log("=== Filtered Results (after applying filters) ==="); console.log("=".repeat(80) + "\n"); const filtered = filterJiraResponse(response); console.log(JSON.stringify(filtered, null, 2)); } else { console.log("No issues found matching the JQL query"); } } catch (error) { console.error("Error fetching JIRA issues:", error.message); if (error.response) { console.error("Response status:", error.response.status); console.error("Response data:", error.response.data); } } } // Export the functions for use in other modules export { filterJiraResponse, extractCommentText, extractTextFromNode }; // Run the test with real JIRA data testCommentExtractor();