agentic-qe 2.3.3 → 2.3.5

package/dist/learning/metrics/LearningMetrics.js

@@ -0,0 +1,754 @@
+"use strict";
+/**
+ * LearningMetrics - Collect and track learning effectiveness metrics
+ *
+ * Aggregates metrics from patterns, dreams, transfers, and experiences
+ * to measure the learning system's effectiveness.
+ *
+ * Part of the Nightly-Learner Phase 3 implementation.
+ *
+ * @version 1.0.0
+ * @module src/learning/metrics/LearningMetrics
+ */
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.LearningMetrics = void 0;
+const better_sqlite3_1 = __importDefault(require("better-sqlite3"));
+const path = __importStar(require("path"));
+const Logger_1 = require("../../utils/Logger");
+/**
+ * LearningMetrics collects and analyzes learning system effectiveness
+ *
+ * @example
+ * ```typescript
+ * const metrics = new LearningMetrics();
+ *
+ * // Get current metrics
+ * const current = await metrics.getCurrentMetrics();
+ * console.log(`Discovery rate: ${current.discoveryRate} patterns/hour`);
+ * console.log(`Transfer success: ${(current.transferSuccessRate * 100).toFixed(1)}%`);
+ *
+ * // Get comprehensive summary with breakdown
+ * const summary = await metrics.getMetricsSummary();
+ * console.log(`Quality trend: ${summary.trends.qualityTrend > 0 ? 'improving' : 'declining'}`);
+ * ```
+ */
+class LearningMetrics {
+    constructor(config) {
+        this.logger = Logger_1.Logger.getInstance();
+        this.config = {
+            dbPath: config?.dbPath || path.join(process.cwd(), '.agentic-qe', 'memory.db'),
+            debug: config?.debug ?? false,
+        };
+        this.db = new better_sqlite3_1.default(this.config.dbPath);
+    }
+    /**
+     * Get current learning metrics
+     */
+    async getCurrentMetrics(periodHours = 24) {
+        const now = Date.now();
+        const periodStart = now - periodHours * 60 * 60 * 1000;
+        // Discovery metrics
+        const discovery = this.calculateDiscoveryMetrics(periodStart, now);
+        // Quality metrics
+        const quality = this.calculateQualityMetrics();
+        // Transfer metrics
+        const transfer = this.calculateTransferMetrics();
+        // Impact metrics
+        const impact = this.calculateImpactMetrics(periodStart, now);
+        // System health
+        const health = this.calculateSystemHealthMetrics(periodStart, now);
+        const metrics = {
+            // Discovery
+            patternsDiscoveredTotal: discovery.total,
+            patternsDiscoveredToday: discovery.today,
+            discoveryRate: discovery.rate,
+            // Quality
+            patternAccuracy: quality.accuracy,
+            insightActionability: quality.actionability,
+            falsePositiveRate: quality.falsePositiveRate,
+            // Transfer
+            transferSuccessRate: transfer.successRate,
+            adoptionRate: transfer.adoptionRate,
+            negativeTransferCount: transfer.negativeTransfers,
+            // Impact
+            taskTimeReduction: impact.timeReduction,
+            coverageImprovement: impact.coverageImprovement,
+            bugDetectionImprovement: impact.bugDetectionImprovement,
+            // System health
+            sleepCycleCompletionRate: health.completionRate,
+            avgCycleDuration: health.avgDuration,
+            errorRate: health.errorRate,
+            // Metadata
+            calculatedAt: new Date(),
+            periodStart: new Date(periodStart),
+            periodEnd: new Date(now),
+        };
+        return metrics;
+    }
+    /**
+     * Get comprehensive metrics summary with breakdown
+     */
+    async getMetricsSummary(periodHours = 24) {
+        const now = Date.now();
+        const periodStart = now - periodHours * 60 * 60 * 1000;
+        const metrics = await this.getCurrentMetrics(periodHours);
+        const breakdown = {
+            discovery: this.getDiscoveryBreakdown(periodStart, now),
+            quality: this.getQualityBreakdown(),
+            transfer: this.getTransferBreakdown(),
+            impact: this.getImpactBreakdown(periodStart, now),
+            system: this.getSystemHealthBreakdown(periodStart, now),
+        };
+        const trends = this.calculateTrends(periodStart, now);
+        return {
+            metrics,
+            breakdown,
+            trends,
+        };
+    }
+    /**
+     * Calculate discovery metrics
+     */
+    calculateDiscoveryMetrics(periodStart, periodEnd) {
+        try {
+            // Total patterns ever discovered
+            const totalRow = this.db.prepare(`
+        SELECT COUNT(*) as count FROM patterns
+      `).get();
+            // Patterns discovered today (last 24 hours)
+            const dayStart = periodEnd - 24 * 60 * 60 * 1000;
+            const todayRow = this.db.prepare(`
+        SELECT COUNT(*) as count FROM patterns
+        WHERE created_at >= ?
+      `).get(dayStart);
+            // Patterns in period
+            const periodRow = this.db.prepare(`
+        SELECT COUNT(*) as count FROM patterns
+        WHERE created_at >= ? AND created_at <= ?
+      `).get(periodStart, periodEnd);
+            const periodHours = (periodEnd - periodStart) / (1000 * 60 * 60);
+            const rate = periodHours > 0 ? periodRow.count / periodHours : 0;
+            return {
+                total: totalRow?.count || 0,
+                today: todayRow?.count || 0,
+                rate,
+            };
+        }
+        catch (error) {
+            if (this.config.debug) {
+                this.logger.debug('[LearningMetrics] Discovery metrics calculation error', { error });
+            }
+            return { total: 0, today: 0, rate: 0 };
+        }
+    }
+    /**
+     * Calculate quality metrics
+     */
+    calculateQualityMetrics() {
+        try {
+            // Pattern accuracy (avg confidence)
+            const confidenceRow = this.db.prepare(`
+        SELECT AVG(confidence) as avg FROM patterns
+        WHERE confidence IS NOT NULL
+      `).get();
+            // Insight actionability (insights applied / total insights)
+            const insightsRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'applied' THEN 1 ELSE 0 END) as applied
+        FROM dream_insights
+      `).get();
+            const actionability = insightsRow?.total > 0
+                ? (insightsRow.applied || 0) / insightsRow.total
+                : 0.5; // Default to neutral
+            // False positive rate (low confidence patterns / total)
+            const falsePositiveRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN confidence < 0.5 THEN 1 ELSE 0 END) as low_confidence
+        FROM patterns
+        WHERE confidence IS NOT NULL
+      `).get();
+            const falsePositiveRate = falsePositiveRow?.total > 0
+                ? (falsePositiveRow.low_confidence || 0) / falsePositiveRow.total
+                : 0;
+            return {
+                accuracy: confidenceRow?.avg || 0.7,
+                actionability,
+                falsePositiveRate,
+            };
+        }
+        catch (error) {
+            if (this.config.debug) {
+                this.logger.debug('[LearningMetrics] Quality metrics calculation error', { error });
+            }
+            return { accuracy: 0.7, actionability: 0.5, falsePositiveRate: 0 };
+        }
+    }
+    /**
+     * Calculate transfer metrics
+     */
+    calculateTransferMetrics() {
+        try {
+            // Transfer success rate from registry
+            const transferRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'active' AND validation_passed = 1 THEN 1 ELSE 0 END) as successful
+        FROM transfer_registry
+      `).get();
+            const successRate = transferRow?.total > 0
+                ? (transferRow.successful || 0) / transferRow.total
+                : 0;
+            // Adoption rate (transferred patterns that were actually used)
+            // We look for patterns in captured_experiences that were transferred
+            const adoptionRow = this.db.prepare(`
+        SELECT
+          COUNT(DISTINCT tr.pattern_id) as transferred,
+          COUNT(DISTINCT ce.id) as used
+        FROM transfer_registry tr
+        LEFT JOIN captured_experiences ce ON
+          JSON_EXTRACT(ce.context, '$.patterns_used') LIKE '%' || tr.pattern_id || '%'
+        WHERE tr.status = 'active'
+      `).get();
+            const adoptionRate = adoptionRow?.transferred > 0
+                ? (adoptionRow.used || 0) / adoptionRow.transferred
+                : 0;
+            // Negative transfers (patterns transferred but led to failures)
+            const negativeRow = this.db.prepare(`
+        SELECT COUNT(DISTINCT tr.id) as count
+        FROM transfer_registry tr
+        JOIN captured_experiences ce ON
+          JSON_EXTRACT(ce.context, '$.patterns_used') LIKE '%' || tr.pattern_id || '%'
+        WHERE tr.status = 'active'
+          AND JSON_EXTRACT(ce.execution, '$.success') = 0
+      `).get();
+            return {
+                successRate,
+                adoptionRate,
+                negativeTransfers: negativeRow?.count || 0,
+            };
+        }
+        catch (error) {
+            if (this.config.debug) {
+                this.logger.debug('[LearningMetrics] Transfer metrics calculation error', { error });
+            }
+            return { successRate: 0, adoptionRate: 0, negativeTransfers: 0 };
+        }
+    }
+    /**
+     * Calculate impact metrics
+     */
+    calculateImpactMetrics(periodStart, periodEnd) {
+        try {
+            // Task time reduction (compare executions with/without patterns)
+            const timeRow = this.db.prepare(`
+        SELECT
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) > 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as with_patterns,
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) = 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as without_patterns
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at <= ?
+          AND JSON_EXTRACT(execution, '$.success') = 1
+      `).get(periodStart, periodEnd);
+            const timeReduction = (timeRow?.without_patterns && timeRow?.with_patterns)
+                ? ((timeRow.without_patterns - timeRow.with_patterns) / timeRow.without_patterns) * 100
+                : 0;
+            // Coverage improvement (would need coverage data - placeholder calculation)
+            const coverageRow = this.db.prepare(`
+        SELECT
+          AVG(CASE WHEN created_at < ? THEN JSON_EXTRACT(outcome, '$.coverage') END) as before,
+          AVG(CASE WHEN created_at >= ? THEN JSON_EXTRACT(outcome, '$.coverage') END) as after
+        FROM captured_experiences
+        WHERE JSON_EXTRACT(outcome, '$.coverage') IS NOT NULL
+      `).get(periodStart, periodStart);
+            const coverageImprovement = (coverageRow?.before && coverageRow?.after)
+                ? ((coverageRow.after - coverageRow.before) / coverageRow.before) * 100
+                : 0;
+            // Bug detection improvement (issues found before/after pattern adoption)
+            const bugRow = this.db.prepare(`
+        SELECT
+          SUM(CASE WHEN created_at < ? THEN JSON_EXTRACT(outcome, '$.bugs_found') END) as before,
+          SUM(CASE WHEN created_at >= ? THEN JSON_EXTRACT(outcome, '$.bugs_found') END) as after
+        FROM captured_experiences
+        WHERE JSON_EXTRACT(outcome, '$.bugs_found') IS NOT NULL
+      `).get(periodStart, periodStart);
+            const bugDetectionImprovement = (bugRow?.before && bugRow?.after && bugRow.before > 0)
+                ? ((bugRow.after - bugRow.before) / bugRow.before) * 100
+                : 0;
+            return {
+                timeReduction,
+                coverageImprovement,
+                bugDetectionImprovement,
+            };
+        }
+        catch (error) {
+            if (this.config.debug) {
+                this.logger.debug('[LearningMetrics] Impact metrics calculation error', { error });
+            }
+            return { timeReduction: 0, coverageImprovement: 0, bugDetectionImprovement: 0 };
+        }
+    }
+    /**
+     * Calculate system health metrics
+     */
+    calculateSystemHealthMetrics(periodStart, periodEnd) {
+        try {
+            // Sleep cycle completion rate
+            const cycleRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'completed' THEN 1 ELSE 0 END) as completed
+        FROM dream_cycles
+        WHERE start_time >= ? AND start_time <= ?
+      `).get(periodStart, periodEnd);
+            const completionRate = cycleRow?.total > 0
+                ? (cycleRow.completed || 0) / cycleRow.total
+                : 0;
+            // Average cycle duration
+            const durationRow = this.db.prepare(`
+        SELECT AVG(duration) as avg
+        FROM dream_cycles
+        WHERE status = 'completed'
+          AND start_time >= ? AND start_time <= ?
+      `).get(periodStart, periodEnd);
+            // Error rate (failed executions / total executions)
+            const errorRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN JSON_EXTRACT(execution, '$.success') = 0 THEN 1 ELSE 0 END) as errors
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at <= ?
+      `).get(periodStart, periodEnd);
+            const errorRate = errorRow?.total > 0
+                ? (errorRow.errors || 0) / errorRow.total
+                : 0;
+            return {
+                completionRate,
+                avgDuration: durationRow?.avg || 0,
+                errorRate,
+            };
+        }
+        catch (error) {
+            if (this.config.debug) {
+                this.logger.debug('[LearningMetrics] System health metrics calculation error', { error });
+            }
+            return { completionRate: 0, avgDuration: 0, errorRate: 0 };
+        }
+    }
+    /**
+     * Get detailed discovery breakdown
+     */
+    getDiscoveryBreakdown(periodStart, periodEnd) {
+        try {
+            const now = Date.now();
+            const dayAgo = now - 24 * 60 * 60 * 1000;
+            const weekAgo = now - 7 * 24 * 60 * 60 * 1000;
+            const monthAgo = now - 30 * 24 * 60 * 60 * 1000;
+            const totalRow = this.db.prepare(`SELECT COUNT(*) as count FROM patterns`).get();
+            const todayRow = this.db.prepare(`SELECT COUNT(*) as count FROM patterns WHERE created_at >= ?`).get(dayAgo);
+            const weekRow = this.db.prepare(`SELECT COUNT(*) as count FROM patterns WHERE created_at >= ?`).get(weekAgo);
+            const monthRow = this.db.prepare(`SELECT COUNT(*) as count FROM patterns WHERE created_at >= ?`).get(monthAgo);
+            const avgPerDay = (monthRow?.count || 0) / 30;
+            // Average patterns per dream cycle
+            const cycleRow = this.db.prepare(`
+        SELECT AVG(insights_generated) as avg
+        FROM dream_cycles
+        WHERE status = 'completed'
+      `).get();
+            return {
+                totalPatterns: totalRow?.count || 0,
+                todayPatterns: todayRow?.count || 0,
+                weekPatterns: weekRow?.count || 0,
+                monthPatterns: monthRow?.count || 0,
+                avgPatternsPerDay: avgPerDay,
+                avgPatternsPerCycle: cycleRow?.avg || 0,
+            };
+        }
+        catch (error) {
+            return {
+                totalPatterns: 0,
+                todayPatterns: 0,
+                weekPatterns: 0,
+                monthPatterns: 0,
+                avgPatternsPerDay: 0,
+                avgPatternsPerCycle: 0,
+            };
+        }
+    }
+    /**
+     * Get detailed quality breakdown
+     */
+    getQualityBreakdown() {
+        try {
+            const confidenceRow = this.db.prepare(`
+        SELECT
+          SUM(CASE WHEN confidence >= 0.8 THEN 1 ELSE 0 END) as high,
+          SUM(CASE WHEN confidence >= 0.5 AND confidence < 0.8 THEN 1 ELSE 0 END) as medium,
+          SUM(CASE WHEN confidence < 0.5 THEN 1 ELSE 0 END) as low,
+          AVG(confidence) as avg
+        FROM patterns
+        WHERE confidence IS NOT NULL
+      `).get();
+            const insightRow = this.db.prepare(`
+        SELECT
+          SUM(CASE WHEN status = 'applied' THEN 1 ELSE 0 END) as applied,
+          SUM(CASE WHEN status = 'pending' THEN 1 ELSE 0 END) as pending,
+          SUM(CASE WHEN status = 'rejected' THEN 1 ELSE 0 END) as rejected
+        FROM dream_insights
+      `).get();
+            return {
+                highConfidencePatterns: confidenceRow?.high || 0,
+                mediumConfidencePatterns: confidenceRow?.medium || 0,
+                lowConfidencePatterns: confidenceRow?.low || 0,
+                avgConfidence: confidenceRow?.avg || 0.7,
+                insightsApplied: insightRow?.applied || 0,
+                insightsPending: insightRow?.pending || 0,
+                insightsRejected: insightRow?.rejected || 0,
+            };
+        }
+        catch (error) {
+            return {
+                highConfidencePatterns: 0,
+                mediumConfidencePatterns: 0,
+                lowConfidencePatterns: 0,
+                avgConfidence: 0.7,
+                insightsApplied: 0,
+                insightsPending: 0,
+                insightsRejected: 0,
+            };
+        }
+    }
+    /**
+     * Get detailed transfer breakdown
+     */
+    getTransferBreakdown() {
+        try {
+            const transferRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'active' THEN 1 ELSE 0 END) as successful,
+          AVG(compatibility_score) as avg_score
+        FROM transfer_registry
+      `).get();
+            const pairRows = this.db.prepare(`
+        SELECT
+          source_agent,
+          target_agent,
+          SUM(CASE WHEN status = 'active' THEN 1 ELSE 0 END) as success,
+          SUM(CASE WHEN status != 'active' THEN 1 ELSE 0 END) as fail
+        FROM transfer_registry
+        GROUP BY source_agent, target_agent
+      `).all();
+            const transfersByAgentPair = new Map();
+            for (const row of pairRows) {
+                const key = `${row.source_agent}->${row.target_agent}`;
+                transfersByAgentPair.set(key, {
+                    success: row.success || 0,
+                    fail: row.fail || 0,
+                });
+            }
+            // Count patterns used post-transfer
+            const usageRow = this.db.prepare(`
+        SELECT COUNT(DISTINCT pattern_id) as count
+        FROM transfer_registry tr
+        WHERE EXISTS (
+          SELECT 1 FROM captured_experiences ce
+          WHERE JSON_EXTRACT(ce.context, '$.patterns_used') LIKE '%' || tr.pattern_id || '%'
+        )
+      `).get();
+            return {
+                totalTransfers: transferRow?.total || 0,
+                successfulTransfers: transferRow?.successful || 0,
+                failedTransfers: (transferRow?.total || 0) - (transferRow?.successful || 0),
+                avgCompatibilityScore: transferRow?.avg_score || 0,
+                transfersByAgentPair,
+                patternsUsedPostTransfer: usageRow?.count || 0,
+            };
+        }
+        catch (error) {
+            return {
+                totalTransfers: 0,
+                successfulTransfers: 0,
+                failedTransfers: 0,
+                avgCompatibilityScore: 0,
+                transfersByAgentPair: new Map(),
+                patternsUsedPostTransfer: 0,
+            };
+        }
+    }
+    /**
+     * Get detailed impact breakdown
+     */
+    getImpactBreakdown(periodStart, periodEnd) {
+        try {
+            const taskRow = this.db.prepare(`
+        SELECT
+          SUM(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) > 0 THEN 1 ELSE 0 END) as with_patterns,
+          SUM(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) = 0 THEN 1 ELSE 0 END) as without_patterns,
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) > 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as time_with,
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) = 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as time_without
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at <= ?
+      `).get(periodStart, periodEnd);
+            const coverageRow = this.db.prepare(`
+        SELECT
+          AVG(CASE WHEN created_at < ? THEN JSON_EXTRACT(outcome, '$.coverage') END) as before,
+          AVG(CASE WHEN created_at >= ? THEN JSON_EXTRACT(outcome, '$.coverage') END) as after
+        FROM captured_experiences
+        WHERE JSON_EXTRACT(outcome, '$.coverage') IS NOT NULL
+      `).get(periodStart, periodStart);
+            const bugRow = this.db.prepare(`
+        SELECT
+          SUM(CASE WHEN created_at < ? THEN JSON_EXTRACT(outcome, '$.bugs_found') END) as before,
+          SUM(CASE WHEN created_at >= ? THEN JSON_EXTRACT(outcome, '$.bugs_found') END) as after
+        FROM captured_experiences
+        WHERE JSON_EXTRACT(outcome, '$.bugs_found') IS NOT NULL
+      `).get(periodStart, periodStart);
+            const timeReductionMs = (taskRow?.time_without && taskRow?.time_with)
+                ? taskRow.time_without - taskRow.time_with
+                : 0;
+            return {
+                tasksWithPatterns: taskRow?.with_patterns || 0,
+                tasksWithoutPatterns: taskRow?.without_patterns || 0,
+                avgTaskTimeWithPatterns: taskRow?.time_with || 0,
+                avgTaskTimeWithoutPatterns: taskRow?.time_without || 0,
+                timeReductionMs,
+                coverageBefore: coverageRow?.before || 0,
+                coverageAfter: coverageRow?.after || 0,
+                bugsDetectedBefore: bugRow?.before || 0,
+                bugsDetectedAfter: bugRow?.after || 0,
+            };
+        }
+        catch (error) {
+            return {
+                tasksWithPatterns: 0,
+                tasksWithoutPatterns: 0,
+                avgTaskTimeWithPatterns: 0,
+                avgTaskTimeWithoutPatterns: 0,
+                timeReductionMs: 0,
+                coverageBefore: 0,
+                coverageAfter: 0,
+                bugsDetectedBefore: 0,
+                bugsDetectedAfter: 0,
+            };
+        }
+    }
+    /**
+     * Get detailed system health breakdown
+     */
+    getSystemHealthBreakdown(periodStart, periodEnd) {
+        try {
+            const cycleRow = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'completed' THEN 1 ELSE 0 END) as completed,
+          SUM(CASE WHEN status = 'failed' THEN 1 ELSE 0 END) as failed,
+          SUM(CASE WHEN status = 'interrupted' THEN 1 ELSE 0 END) as interrupted,
+          AVG(duration) as avg_duration,
+          MIN(duration) as min_duration,
+          MAX(duration) as max_duration
+        FROM dream_cycles
+        WHERE start_time >= ? AND start_time <= ?
+      `).get(periodStart, periodEnd);
+            const errorRow = this.db.prepare(`
+        SELECT COUNT(*) as count
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at <= ?
+          AND JSON_EXTRACT(execution, '$.success') = 0
+      `).get(periodStart, periodEnd);
+            // Group errors by type
+            const errorTypeRows = this.db.prepare(`
+        SELECT
+          JSON_EXTRACT(context, '$.errors_encountered[0]') as error_type,
+          COUNT(*) as count
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at <= ?
+          AND JSON_EXTRACT(execution, '$.success') = 0
+        GROUP BY error_type
+      `).all(periodStart, periodEnd);
+            const errorsByType = new Map();
+            for (const row of errorTypeRows) {
+                if (row.error_type) {
+                    errorsByType.set(row.error_type, row.count || 0);
+                }
+            }
+            return {
+                totalCycles: cycleRow?.total || 0,
+                completedCycles: cycleRow?.completed || 0,
+                failedCycles: cycleRow?.failed || 0,
+                interruptedCycles: cycleRow?.interrupted || 0,
+                avgCycleDurationMs: cycleRow?.avg_duration || 0,
+                minCycleDurationMs: cycleRow?.min_duration || 0,
+                maxCycleDurationMs: cycleRow?.max_duration || 0,
+                totalErrors: errorRow?.count || 0,
+                errorsByType,
+            };
+        }
+        catch (error) {
+            return {
+                totalCycles: 0,
+                completedCycles: 0,
+                failedCycles: 0,
+                interruptedCycles: 0,
+                avgCycleDurationMs: 0,
+                minCycleDurationMs: 0,
+                maxCycleDurationMs: 0,
+                totalErrors: 0,
+                errorsByType: new Map(),
+            };
+        }
+    }
+    /**
+     * Calculate trends over time
+     */
+    calculateTrends(periodStart, periodEnd) {
+        try {
+            const periodDuration = periodEnd - periodStart;
+            const midpoint = periodStart + periodDuration / 2;
+            // Discovery trend (compare first half vs second half)
+            const discoveryFirstHalf = this.db.prepare(`
+        SELECT COUNT(*) as count FROM patterns
+        WHERE created_at >= ? AND created_at < ?
+      `).get(periodStart, midpoint);
+            const discoverySecondHalf = this.db.prepare(`
+        SELECT COUNT(*) as count FROM patterns
+        WHERE created_at >= ? AND created_at <= ?
+      `).get(midpoint, periodEnd);
+            const discoveryTrend = (discoveryFirstHalf?.count || 0) > 0
+                ? ((discoverySecondHalf?.count || 0) - (discoveryFirstHalf?.count || 0)) / (discoveryFirstHalf.count || 1)
+                : 0;
+            // Quality trend (compare average confidence)
+            const qualityFirstHalf = this.db.prepare(`
+        SELECT AVG(confidence) as avg FROM patterns
+        WHERE created_at >= ? AND created_at < ?
+      `).get(periodStart, midpoint);
+            const qualitySecondHalf = this.db.prepare(`
+        SELECT AVG(confidence) as avg FROM patterns
+        WHERE created_at >= ? AND created_at <= ?
+      `).get(midpoint, periodEnd);
+            const qualityTrend = (qualityFirstHalf?.avg || 0) > 0
+                ? ((qualitySecondHalf?.avg || 0) - (qualityFirstHalf?.avg || 0)) / (qualityFirstHalf.avg || 1)
+                : 0;
+            // Transfer trend (compare success rates)
+            const transferFirstHalf = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'active' THEN 1 ELSE 0 END) as successful
+        FROM transfer_registry
+        WHERE transferred_at >= ? AND transferred_at < ?
+      `).get(periodStart, midpoint);
+            const transferSecondHalf = this.db.prepare(`
+        SELECT
+          COUNT(*) as total,
+          SUM(CASE WHEN status = 'active' THEN 1 ELSE 0 END) as successful
+        FROM transfer_registry
+        WHERE transferred_at >= ? AND transferred_at <= ?
+      `).get(midpoint, periodEnd);
+            const transferRateFirst = (transferFirstHalf?.total || 0) > 0
+                ? (transferFirstHalf.successful || 0) / transferFirstHalf.total
+                : 0;
+            const transferRateSecond = (transferSecondHalf?.total || 0) > 0
+                ? (transferSecondHalf.successful || 0) / transferSecondHalf.total
+                : 0;
+            const transferTrend = transferRateFirst > 0
+                ? (transferRateSecond - transferRateFirst) / transferRateFirst
+                : 0;
+            // Impact trend (compare task time reduction)
+            const impactFirstHalf = this.db.prepare(`
+        SELECT
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) > 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as with_patterns,
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) = 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as without_patterns
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at < ?
+      `).get(periodStart, midpoint);
+            const impactSecondHalf = this.db.prepare(`
+        SELECT
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) > 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as with_patterns,
+          AVG(CASE WHEN JSON_ARRAY_LENGTH(JSON_EXTRACT(context, '$.patterns_used')) = 0
+            THEN JSON_EXTRACT(execution, '$.duration') END) as without_patterns
+        FROM captured_experiences
+        WHERE created_at >= ? AND created_at <= ?
+      `).get(midpoint, periodEnd);
+            const reductionFirst = (impactFirstHalf?.without_patterns && impactFirstHalf?.with_patterns)
+                ? (impactFirstHalf.without_patterns - impactFirstHalf.with_patterns) / impactFirstHalf.without_patterns
+                : 0;
+            const reductionSecond = (impactSecondHalf?.without_patterns && impactSecondHalf?.with_patterns)
+                ? (impactSecondHalf.without_patterns - impactSecondHalf.with_patterns) / impactSecondHalf.without_patterns
+                : 0;
+            const impactTrend = reductionFirst > 0
+                ? (reductionSecond - reductionFirst) / reductionFirst
+                : 0;
+            // Normalize trends to -1 to 1 range
+            const normalize = (value) => Math.max(-1, Math.min(1, value));
+            return {
+                discoveryTrend: normalize(discoveryTrend),
+                qualityTrend: normalize(qualityTrend),
+                transferTrend: normalize(transferTrend),
+                impactTrend: normalize(impactTrend),
+            };
+        }
+        catch (error) {
+            if (this.config.debug) {
+                this.logger.debug('[LearningMetrics] Trend calculation error', { error });
+            }
+            return {
+                discoveryTrend: 0,
+                qualityTrend: 0,
+                transferTrend: 0,
+                impactTrend: 0,
+            };
+        }
+    }
+    /**
+     * Close database connection
+     */
+    close() {
+        this.db.close();
+    }
+}
+exports.LearningMetrics = LearningMetrics;
+exports.default = LearningMetrics;
+//# sourceMappingURL=LearningMetrics.js.map