Quality Dimension Generator

import { TaskAnalysis, TimeContext, QualityDimensionsResponse, QualityDimension } from './types.js'; /** * Quality Evaluation Dimension Generator * Supports configurable dimension count and expected score */ export class QualityDimensionGenerator { private dimensionCount: number; private expectedScore: number; constructor(config: { dimensionCount: number; expectedScore: number } = { dimensionCount: 5, expectedScore: 8 }) { this.dimensionCount = config.dimensionCount; this.expectedScore = config.expectedScore; } /** * Generate dimension scoring prompt using configured dimension count and expected score */ public async generateDimensionsPrompt( task: TaskAnalysis, timeContext: TimeContext ): Promise<string> { // Use instance configuration const dimensionCount = this.dimensionCount; const expectedScore = this.expectedScore; return `Please analyze the following task and generate ${dimensionCount} evaluation dimensions in JSON format: ## 📋 Task Information - **Core Task**: ${task.coreTask} - **Task Type**: ${task.taskType} - **Complexity**: ${task.complexity}/5 - **Domain**: ${task.domain} - **Key Elements**: ${task.keyElements.join(', ')} - **Objectives**: ${task.objectives.join(', ')} ## ⏰ Time Context - **Current Time**: ${timeContext.formattedTime} - **Year**: ${timeContext.year}, Month: ${timeContext.month} ## 🎯 Quality Target - **Expected Score**: ${expectedScore}/10 points - **Total Score Calculation**: Average of all ${dimensionCount} dimension scores Please analyze according to the following requirements and output the result in JSON format: \`\`\`json { "expectedScore": ${expectedScore}, "scoreCalculation": "Average of all ${dimensionCount} dimension scores", "dimensions": [ { "name": "Dimension name (3-15 characters, specific and clear)", "description": "One sentence description of what this dimension evaluates", "importance": "One sentence explaining why this dimension is critical for the task", "scoring": { "10": "Specific criteria for excellent performance (measurable and actionable)", "8": "Specific criteria for good performance (measurable and actionable)", "6": "Specific criteria for acceptable performance (measurable and actionable)" } }, { "name": "Second dimension name", "description": "Description of second dimension", "importance": "Why this dimension matters", "scoring": { "10": "Excellent criteria for dimension 2", "8": "Good criteria for dimension 2", "6": "Acceptable criteria for dimension 2" } } // ... continue for all ${dimensionCount} dimensions ] } \`\`\` ## ✅ Analysis Requirements 1. **Dimension Count**: Generate exactly ${dimensionCount} dimensions that comprehensively cover the task 2. **Dimension Names**: Use concise, professional names (3-15 characters) that clearly indicate what is being evaluated 3. **Descriptions**: One clear sentence explaining the scope of each dimension's evaluation 4. **Importance**: Explain why each dimension is essential for achieving task success 5. **Scoring Criteria**: Each score level (6, 8, 10) must have specific, measurable, actionable criteria 6. **Quality Standards**: Design criteria so that achieving an average of ${expectedScore}/10 across all dimensions represents realistic excellence for this task 7. **Domain Relevance**: All dimensions must be appropriate for the "${task.domain}" domain 8. **Comprehensive Coverage**: The ${dimensionCount} dimensions together should evaluate all critical aspects of the task ## ⚠️ JSON Format Requirements - Ensure valid JSON syntax with proper quotes and commas - Root object must include expectedScore (number) and scoreCalculation (string) fields - Each dimension object must include all 4 fields: name, description, importance, scoring - Scoring object must have exactly 3 levels: "6", "8", "10" - All text should be professional and specific to the task domain - Generate complete and valid JSON that can be directly parsed ## 🎯 Final Reminder Generate exactly ${dimensionCount} complete dimensions that together provide a comprehensive evaluation framework for: "${task.coreTask}" **Important**: Each dimension must include all 4 required fields (name, description, importance, scoring with 6/8/10 levels). Please ensure the JSON is complete, valid, and directly usable for evaluation purposes.`; } /** * Parse LLM-returned quality dimensions result */ public parseQualityDimensionsResult(llmResponse: string): QualityDimensionsResponse { // Extract JSON part const jsonMatch = llmResponse.match(/```json\s*([\s\S]*?)\s*```/); if (!jsonMatch) { throw new Error('JSON format quality dimensions result not found'); } const result = JSON.parse(jsonMatch[1]); // Basic validation if (!result.dimensions || !Array.isArray(result.dimensions)) { throw new Error('Missing or invalid dimensions array'); } if (typeof result.expectedScore !== 'number') { throw new Error('Missing or invalid expectedScore field'); } if (typeof result.scoreCalculation !== 'string') { throw new Error('Missing or invalid scoreCalculation field'); } return result as QualityDimensionsResponse; } /** * Parse and validate LLM-returned quality dimensions result */ public parseDimensionsResult(llmResponse: string): QualityDimensionsResponse { // Extract JSON part const jsonMatch = llmResponse.match(/```json\s*([\s\S]*?)\s*```/); if (!jsonMatch) { throw new Error('JSON format dimensions result not found'); } const result = JSON.parse(jsonMatch[1]); // Validate root structure if (!result.dimensions || !Array.isArray(result.dimensions)) { throw new Error('Invalid structure: missing or invalid dimensions array'); } if (typeof result.expectedScore !== 'number') { throw new Error('Missing or invalid expectedScore field'); } if (typeof result.scoreCalculation !== 'string') { throw new Error('Missing or invalid scoreCalculation field'); } // Validate dimension count if (result.dimensions.length !== this.dimensionCount) { throw new Error(`Expected ${this.dimensionCount} dimensions, got ${result.dimensions.length}`); } // Validate each dimension result.dimensions.forEach((dimension: any, index: number) => { this.validateDimension(dimension, index); }); return { expectedScore: result.expectedScore, scoreCalculation: result.scoreCalculation, dimensions: result.dimensions as QualityDimension[] }; } /** * Validate a single dimension object */ private validateDimension(dimension: any, index: number): void { const requiredFields = ['name', 'description', 'importance', 'scoring']; // Check required fields for (const field of requiredFields) { if (!dimension[field]) { throw new Error(`Dimension ${index + 1}: missing required field '${field}'`); } } // Validate name length if (dimension.name.length < 3 || dimension.name.length > 15) { throw new Error(`Dimension ${index + 1}: name should be 3-15 characters, got ${dimension.name.length}`); } // Validate scoring structure if (!dimension.scoring || typeof dimension.scoring !== 'object') { throw new Error(`Dimension ${index + 1}: invalid scoring object`); } const requiredScores = ['6', '8', '10']; for (const score of requiredScores) { if (!dimension.scoring[score] || typeof dimension.scoring[score] !== 'string') { throw new Error(`Dimension ${index + 1}: missing or invalid scoring criteria for level ${score}`); } } // Check for extra scoring levels const actualScores = Object.keys(dimension.scoring); const extraScores = actualScores.filter(score => !requiredScores.includes(score)); if (extraScores.length > 0) { throw new Error(`Dimension ${index + 1}: unexpected scoring levels: ${extraScores.join(', ')}`); } } }