Jira MCP Toolset

/** * Acceptance criteria evaluation functionality * * This file implements the logic for analyzing and evaluating the quality * and completeness of acceptance criteria within Jira issues, detecting * missing elements and assessing overall criteria quality. */ import type { CategoryCheckResult } from './completenessEvaluation.types' import type { JiraIssue } from '../../../types/issue.types' /** * Checks if acceptance criteria are present and of good quality in a Jira issue */ export function checkAcceptanceCriteria(allText: string, issue: JiraIssue): CategoryCheckResult { const acceptanceCriteriaPatterns = [ /acceptance criteria/i, /given.*when.*then/i, /success criteria/i, /definition of done/i, /acceptance test/i, /should (be able to|allow)/i, ] // Look for acceptance criteria in the text const hasPatternMatch = acceptanceCriteriaPatterns.some((pattern) => pattern.test(allText)) // Special case: Check if there's a dedicated custom field for acceptance criteria // This is just an example and would need to be adapted to the actual Jira instance const hasCustomField = Boolean(issue.fields['customfield_10101']) // Determine presence and quality const isPresent = hasPatternMatch || hasCustomField let quality: 'absent' | 'partial' | 'complete' = 'absent' if (isPresent) { // Basic heuristic: If we found multiple patterns, it's more likely to be complete // More sophisticated analysis could look at structure and word count const matchCount = acceptanceCriteriaPatterns.filter((pattern) => pattern.test(allText)).length quality = matchCount > 2 ? 'complete' : 'partial' } return { missing: isPresent ? [] : ['Acceptance criteria not found'], present: isPresent, quality, } }