diff --git a/services/mcp-server/src/modules/a2a/engagement-pipeline.ts b/services/mcp-server/src/modules/a2a/engagement-pipeline.ts
new file mode 100644
index 00000000..ddcd1bb5
--- /dev/null
+++ b/services/mcp-server/src/modules/a2a/engagement-pipeline.ts
@@ -0,0 +1,240 @@
+/**
+ * EngagementAgent — A2A pipeline for NomGap user engagement recovery.
+ *
+ * Agent roster (3 steps):
+ *   1. StreakRiskDetectorAgent  — query telemetry for streak_risk events; cross-ref fasting stats
+ *   2. EngagementTriggerAgent   — fire streak_risk push for at-risk users; escalate to weekly_digest if streak = 0
+ *   3. EngagementReportAgent    — assemble per-user engagement action report
+ *
+ * MCP tools:
+ *   nomgap.engagement.recover(userId?, from?, to?) — run pipeline for one or all at-risk users
+ */
+
+import { randomUUID } from 'node:crypto';
+import { z } from 'zod';
+import { registerTool } from '../tools/registry.js';
+import type { McpToolRequest } from '../tools/types.js';
+import { telemetryQuery } from '../../lib/platform-client.js';
+import { nomgapFastingGetStats, nomgapPushFire } from '../../lib/nomgap-client.js';
+
+// ── Types ──────────────────────────────────────────────────────────────────────
+
+type EngagementAction = 'streak_risk' | 'weekly_digest' | 'none';
+
+interface RiskSignal {
+  userId: string;
+  recentRiskEvents: number;
+  currentStreak: number | null;
+  longestFast: number | null;
+}
+
+interface TriggerResult {
+  userId: string;
+  action: EngagementAction;
+  fired: boolean;
+  skipReason?: string;
+  error?: string;
+}
+
+export interface EngagementReport {
+  runId: string;
+  productId: 'nomgap';
+  usersScanned: number;
+  atRiskCount: number;
+  zeroStreakCount: number;
+  fired: number;
+  skipped: number;
+  failed: number;
+  perUser: TriggerResult[];
+  summary: string;
+  generatedAt: string;
+}
+
+// ── Step 1: StreakRiskDetectorAgent ────────────────────────────────────────────
+
+async function detectAtRiskUsers(
+  userIdFilter: string | null,
+  fromTime: string,
+  toTime: string,
+  opts: { token?: string; requestId?: string }
+): Promise<RiskSignal[]> {
+  let riskUserIds: string[] = [];
+
+  try {
+    const result = await telemetryQuery(
+      {
+        productId: 'nomgap',
+        eventType: 'streak_risk',
+        from: fromTime,
+        to: toTime,
+        limit: 50,
+      },
+      { ...opts, productId: 'nomgap' }
+    );
+
+    const events = result.events as Array<Record<string, unknown>>;
+    const seen = new Set<string>();
+    for (const event of events) {
+      const uid = (event['userId'] as string) || (event['anonymousId'] as string);
+      if (uid && !seen.has(uid)) {
+        seen.add(uid);
+        riskUserIds.push(uid);
+      }
+    }
+  } catch {
+    // best-effort
+  }
+
+  // If caller specified a single userId, target only that user
+  if (userIdFilter) {
+    if (!riskUserIds.includes(userIdFilter)) riskUserIds = [userIdFilter];
+    else riskUserIds = [userIdFilter];
+  }
+
+  const signals: RiskSignal[] = [];
+  for (const userId of riskUserIds) {
+    let currentStreak: number | null = null;
+    let longestFast: number | null = null;
+
+    try {
+      const stats = await nomgapFastingGetStats({ token: opts.token, requestId: opts.requestId });
+      const s = stats as Record<string, unknown>;
+      currentStreak = typeof s['currentStreak'] === 'number' ? s['currentStreak'] : null;
+      longestFast = typeof s['longestFastHours'] === 'number' ? s['longestFastHours'] : null;
+    } catch {
+      // best-effort
+    }
+
+    signals.push({
+      userId,
+      recentRiskEvents: riskUserIds.filter(id => id === userId).length,
+      currentStreak,
+      longestFast,
+    });
+  }
+
+  return signals;
+}
+
+// ── Step 2: EngagementTriggerAgent ────────────────────────────────────────────
+
+async function fireTrigger(
+  signal: RiskSignal,
+  opts: { token?: string; requestId?: string }
+): Promise<TriggerResult> {
+  const { userId, currentStreak } = signal;
+
+  // Escalate to weekly_digest if streak has dropped to zero
+  const action: EngagementAction = currentStreak === 0 ? 'weekly_digest' : 'streak_risk';
+
+  try {
+    await nomgapPushFire(
+      { type: action, userId },
+      { token: opts.token, requestId: opts.requestId }
+    );
+    return { userId, action, fired: true };
+  } catch (err) {
+    return {
+      userId,
+      action,
+      fired: false,
+      error: err instanceof Error ? err.message : String(err),
+    };
+  }
+}
+
+// ── Step 3: EngagementReportAgent ─────────────────────────────────────────────
+
+function buildEngagementReport(
+  runId: string,
+  signals: RiskSignal[],
+  results: TriggerResult[]
+): EngagementReport {
+  const now = new Date().toISOString();
+  const fired = results.filter(r => r.fired).length;
+  const skipped = results.filter(r => !r.fired && !r.error).length;
+  const failed = results.filter(r => !!r.error).length;
+  const zeroStreakCount = signals.filter(s => s.currentStreak === 0).length;
+
+  const summary =
+    signals.length === 0
+      ? 'No at-risk users detected in the specified window.'
+      : `${fired} engagement push(es) fired for ${signals.length} at-risk user(s). ${zeroStreakCount} escalated to weekly_digest (streak = 0). ${failed} failed.`;
+
+  return {
+    runId,
+    productId: 'nomgap',
+    usersScanned: signals.length,
+    atRiskCount: signals.length,
+    zeroStreakCount,
+    fired,
+    skipped,
+    failed,
+    perUser: results,
+    summary,
+    generatedAt: now,
+  };
+}
+
+// ── Pipeline runner ────────────────────────────────────────────────────────────
+
+async function runEngagementPipeline(
+  userIdFilter: string | null,
+  fromTime: string,
+  toTime: string,
+  req: McpToolRequest
+): Promise<EngagementReport> {
+  const runId = randomUUID();
+  const opts = { token: req.headers.authorization?.slice(7), requestId: req.id };
+
+  req.log.info({ runId, stepId: 'detect', userIdFilter }, 'StreakRiskDetectorAgent start');
+  const signals = await detectAtRiskUsers(userIdFilter, fromTime, toTime, opts);
+  req.log.info(
+    { runId, stepId: 'detect', atRiskCount: signals.length },
+    'StreakRiskDetectorAgent done'
+  );
+
+  req.log.info({ runId, stepId: 'trigger', count: signals.length }, 'EngagementTriggerAgent start');
+  const results: TriggerResult[] = [];
+  for (const signal of signals) {
+    const result = await fireTrigger(signal, opts);
+    results.push(result);
+  }
+  req.log.info(
+    { runId, stepId: 'trigger', fired: results.filter(r => r.fired).length },
+    'EngagementTriggerAgent done'
+  );
+
+  req.log.info({ runId, stepId: 'report' }, 'EngagementReportAgent start');
+  const report = buildEngagementReport(runId, signals, results);
+  req.log.info({ runId, stepId: 'report', summary: report.summary }, 'EngagementReportAgent done');
+
+  return report;
+}
+
+// ── MCP tool registration ─────────────────────────────────────────────────────
+
+registerTool({
+  name: 'nomgap.engagement.recover',
+  description:
+    'A2A pipeline: detects at-risk NomGap users from streak_risk telemetry events, fires streak_risk push notifications (or weekly_digest for zero-streak users), and returns a per-user engagement action report. Optionally target a single userId. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    userId: z
+      .string()
+      .optional()
+      .describe('Target a specific user (omit to process all at-risk users from telemetry)'),
+    from: z
+      .string()
+      .datetime()
+      .optional()
+      .describe('Start of risk signal window (default: 24h ago)'),
+    to: z.string().datetime().optional().describe('End of risk signal window (default: now)'),
+  }),
+  async execute(args, req) {
+    const now = new Date();
+    const toTime = args.to ?? now.toISOString();
+    const fromTime = args.from ?? new Date(now.getTime() - 24 * 60 * 60 * 1000).toISOString();
+    return runEngagementPipeline(args.userId ?? null, fromTime, toTime, req);
+  },
+});
diff --git a/services/mcp-server/src/modules/a2a/memory-curation-pipeline.ts b/services/mcp-server/src/modules/a2a/memory-curation-pipeline.ts
new file mode 100644
index 00000000..71cae270
--- /dev/null
+++ b/services/mcp-server/src/modules/a2a/memory-curation-pipeline.ts
@@ -0,0 +1,225 @@
+/**
+ * MemoryCurationAgent — A2A pipeline for JarvisJr agent memory hygiene.
+ *
+ * Agent roster (3 steps):
+ *   1. AgentInventoryAgent   — list all agents owned by a user
+ *   2. MemoryScanAgent       — per agent: list memories below importance threshold
+ *   3. CurationReportAgent   — prune stale entries + return per-agent curation report
+ *
+ * MCP tools:
+ *   jarvis.memory.curate(importanceThreshold?, dryRun?) — run pipeline across all agents
+ */
+
+import { randomUUID } from 'node:crypto';
+import { z } from 'zod';
+import { registerTool } from '../tools/registry.js';
+import type { McpToolRequest } from '../tools/types.js';
+import {
+  jarvisAgentsList,
+  jarvisMemoryList,
+  jarvisMemoryPrune,
+  type JarvisMemoryDoc,
+} from '../../lib/jarvis-client.js';
+import { config } from '../../lib/config.js';
+
+// ── Types ──────────────────────────────────────────────────────────────────────
+
+interface AgentMemoryScan {
+  agentId: string;
+  agentName: string;
+  totalMemories: number;
+  staleMemories: JarvisMemoryDoc[];
+  expiredMemories: JarvisMemoryDoc[];
+}
+
+interface CurationAction {
+  agentId: string;
+  agentName: string;
+  staleCount: number;
+  expiredCount: number;
+  pruned: number;
+  skipped: boolean;
+}
+
+export interface MemoryCurationReport {
+  runId: string;
+  productId: 'jarvisjr';
+  importanceThreshold: number;
+  dryRun: boolean;
+  agentsScanned: number;
+  totalStaleFound: number;
+  totalExpiredFound: number;
+  totalPruned: number;
+  perAgent: CurationAction[];
+  summary: string;
+  generatedAt: string;
+}
+
+// ── Step 1: AgentInventoryAgent ────────────────────────────────────────────────
+
+async function listAgents(opts: {
+  token?: string;
+  requestId?: string;
+}): Promise<Array<{ id: string; name: string }>> {
+  try {
+    const result = await jarvisAgentsList({ limit: config.QUERY_MAX_LIMIT }, opts);
+    return result.agents.map(a => ({ id: a.id, name: a.name }));
+  } catch {
+    return [];
+  }
+}
+
+// ── Step 2: MemoryScanAgent ────────────────────────────────────────────────────
+
+async function scanAgentMemory(
+  agentId: string,
+  agentName: string,
+  importanceThreshold: number,
+  opts: { token?: string; requestId?: string }
+): Promise<AgentMemoryScan> {
+  try {
+    const result = await jarvisMemoryList(agentId, { limit: config.QUERY_MAX_LIMIT }, opts);
+    const now = new Date();
+
+    const staleMemories = result.memories.filter(m => m.importance < importanceThreshold);
+    const expiredMemories = result.memories.filter(m => m.expiresAt && new Date(m.expiresAt) < now);
+
+    return {
+      agentId,
+      agentName,
+      totalMemories: result.total,
+      staleMemories,
+      expiredMemories,
+    };
+  } catch {
+    return { agentId, agentName, totalMemories: 0, staleMemories: [], expiredMemories: [] };
+  }
+}
+
+// ── Step 3: CurationReportAgent ────────────────────────────────────────────────
+
+async function curateAgentMemory(
+  scan: AgentMemoryScan,
+  dryRun: boolean,
+  opts: { token?: string; requestId?: string }
+): Promise<CurationAction> {
+  const staleCount = scan.staleMemories.length;
+  const expiredCount = scan.expiredMemories.length;
+
+  if (dryRun || (staleCount === 0 && expiredCount === 0)) {
+    return {
+      agentId: scan.agentId,
+      agentName: scan.agentName,
+      staleCount,
+      expiredCount,
+      pruned: 0,
+      skipped: dryRun || (staleCount === 0 && expiredCount === 0),
+    };
+  }
+
+  let pruned = 0;
+  try {
+    const result = await jarvisMemoryPrune(scan.agentId, opts);
+    pruned = result.pruned;
+  } catch {
+    // best-effort
+  }
+
+  return {
+    agentId: scan.agentId,
+    agentName: scan.agentName,
+    staleCount,
+    expiredCount,
+    pruned,
+    skipped: false,
+  };
+}
+
+// ── Pipeline runner ────────────────────────────────────────────────────────────
+
+async function runMemoryCurationPipeline(
+  importanceThreshold: number,
+  dryRun: boolean,
+  req: McpToolRequest
+): Promise<MemoryCurationReport> {
+  const runId = randomUUID();
+  const opts = { token: req.headers.authorization?.slice(7), requestId: req.id };
+
+  req.log.info(
+    { runId, stepId: 'inventory', dryRun, importanceThreshold },
+    'AgentInventoryAgent start'
+  );
+  const agents = await listAgents(opts);
+  req.log.info(
+    { runId, stepId: 'inventory', agentCount: agents.length },
+    'AgentInventoryAgent done'
+  );
+
+  req.log.info({ runId, stepId: 'scan', agentCount: agents.length }, 'MemoryScanAgent start');
+  const scans: AgentMemoryScan[] = [];
+  for (const agent of agents) {
+    const scan = await scanAgentMemory(agent.id, agent.name, importanceThreshold, opts);
+    scans.push(scan);
+  }
+  req.log.info(
+    { runId, stepId: 'scan', totalStale: scans.reduce((s, x) => s + x.staleMemories.length, 0) },
+    'MemoryScanAgent done'
+  );
+
+  req.log.info({ runId, stepId: 'curate', dryRun }, 'CurationReportAgent start');
+  const actions: CurationAction[] = [];
+  for (const scan of scans) {
+    const action = await curateAgentMemory(scan, dryRun, opts);
+    actions.push(action);
+  }
+
+  const totalStaleFound = actions.reduce((s, a) => s + a.staleCount, 0);
+  const totalExpiredFound = actions.reduce((s, a) => s + a.expiredCount, 0);
+  const totalPruned = actions.reduce((s, a) => s + a.pruned, 0);
+
+  const summary = dryRun
+    ? `DRY RUN: ${totalStaleFound} stale + ${totalExpiredFound} expired memories found across ${agents.length} agents. No pruning performed.`
+    : totalPruned > 0
+      ? `Pruned ${totalPruned} entries across ${actions.filter(a => !a.skipped).length} agents (${totalStaleFound} stale, ${totalExpiredFound} expired found).`
+      : `No memories required pruning across ${agents.length} agents.`;
+
+  req.log.info({ runId, stepId: 'curate', totalPruned, summary }, 'CurationReportAgent done');
+
+  return {
+    runId,
+    productId: 'jarvisjr',
+    importanceThreshold,
+    dryRun,
+    agentsScanned: agents.length,
+    totalStaleFound,
+    totalExpiredFound,
+    totalPruned,
+    perAgent: actions,
+    summary,
+    generatedAt: new Date().toISOString(),
+  };
+}
+
+// ── MCP tool registration ─────────────────────────────────────────────────────
+
+registerTool({
+  name: 'jarvis.memory.curate',
+  description:
+    'A2A pipeline: scans all JarvisJr coaching agents, identifies memories below an importance threshold or past their expiry date, prunes stale entries, and returns a per-agent hygiene report. Use dryRun=true to preview without pruning. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    importanceThreshold: z.coerce
+      .number()
+      .min(0)
+      .max(1)
+      .default(0.3)
+      .describe('Memories with importance below this value are considered stale (default 0.3)'),
+    dryRun: z
+      .boolean()
+      .default(false)
+      .describe('If true, scan and report without pruning any memories'),
+  }),
+  async execute(args, req) {
+    return runMemoryCurationPipeline(args.importanceThreshold, args.dryRun, req);
+  },
+});
diff --git a/services/mcp-server/src/modules/a2a/triage-quality-pipeline.ts b/services/mcp-server/src/modules/a2a/triage-quality-pipeline.ts
new file mode 100644
index 00000000..cc37ca57
--- /dev/null
+++ b/services/mcp-server/src/modules/a2a/triage-quality-pipeline.ts
@@ -0,0 +1,252 @@
+/**
+ * TriageQualityAgent — A2A pipeline for MindLyst memory triage quality improvement.
+ *
+ * Agent roster (3 steps):
+ *   1. LowConfidenceCollectorAgent — list memory items, filter for confidenceScore below threshold
+ *   2. RetriageAgent               — re-run extraction on each low-confidence item
+ *   3. TriageQualityReportAgent    — compare old vs new routing, auto-reassign if brainId changed
+ *
+ * MCP tools:
+ *   mindlyst.memory.triageQuality(confidenceThreshold?, dryRun?) — run pipeline
+ */
+
+import { randomUUID } from 'node:crypto';
+import { z } from 'zod';
+import { registerTool } from '../tools/registry.js';
+import type { McpToolRequest } from '../tools/types.js';
+import {
+  mindlystMemoryList,
+  mindlystMemoryRetriage,
+  mindlystMemoryReassign,
+} from '../../lib/mindlyst-client.js';
+import { config } from '../../lib/config.js';
+
+// ── Types ──────────────────────────────────────────────────────────────────────
+
+interface LowConfidenceItem {
+  id: string;
+  confidenceScore: number;
+  currentBrainId: string | null;
+  contentType: string;
+}
+
+interface RetriageResult {
+  itemId: string;
+  oldConfidenceScore: number;
+  newConfidenceScore: number;
+  oldBrainId: string | null;
+  newBrainId: string | null;
+  brainChanged: boolean;
+  reassigned: boolean;
+  error?: string;
+}
+
+export interface TriageQualityReport {
+  runId: string;
+  productId: 'mindlyst';
+  confidenceThreshold: number;
+  dryRun: boolean;
+  totalFetched: number;
+  lowConfidenceCount: number;
+  retriaged: number;
+  improved: number;
+  brainChanges: number;
+  reassigned: number;
+  failed: number;
+  perItem: RetriageResult[];
+  summary: string;
+  generatedAt: string;
+}
+
+// ── Step 1: LowConfidenceCollectorAgent ───────────────────────────────────────
+
+async function collectLowConfidenceItems(
+  confidenceThreshold: number,
+  opts: { token?: string; requestId?: string }
+): Promise<{ items: LowConfidenceItem[]; totalFetched: number }> {
+  try {
+    const result = await mindlystMemoryList({ limit: config.QUERY_MAX_LIMIT }, opts);
+
+    const allItems = result.items;
+    const lowConfidence = allItems
+      .filter(item => item.triageResult.confidenceScore < confidenceThreshold)
+      .map(item => ({
+        id: item.id,
+        confidenceScore: item.triageResult.confidenceScore,
+        currentBrainId: item.brainIds[0] ?? null,
+        contentType: item.triageResult.contentType,
+      }));
+
+    return { items: lowConfidence, totalFetched: allItems.length };
+  } catch {
+    return { items: [], totalFetched: 0 };
+  }
+}
+
+// ── Step 2: RetriageAgent ─────────────────────────────────────────────────────
+
+async function retriageItem(
+  item: LowConfidenceItem,
+  dryRun: boolean,
+  opts: { token?: string; requestId?: string }
+): Promise<RetriageResult> {
+  if (dryRun) {
+    return {
+      itemId: item.id,
+      oldConfidenceScore: item.confidenceScore,
+      newConfidenceScore: item.confidenceScore,
+      oldBrainId: item.currentBrainId,
+      newBrainId: item.currentBrainId,
+      brainChanged: false,
+      reassigned: false,
+    };
+  }
+
+  try {
+    const updated = await mindlystMemoryRetriage(item.id, opts);
+    const newConfidence = updated.triageResult.confidenceScore;
+    const newBrainId = updated.triageResult.suggestedBrainId ?? updated.brainIds[0] ?? null;
+    const brainChanged = newBrainId !== null && newBrainId !== item.currentBrainId;
+
+    let reassigned = false;
+    if (brainChanged && newBrainId) {
+      try {
+        await mindlystMemoryReassign(item.id, newBrainId, opts);
+        reassigned = true;
+      } catch {
+        // best-effort reassignment
+      }
+    }
+
+    return {
+      itemId: item.id,
+      oldConfidenceScore: item.confidenceScore,
+      newConfidenceScore: newConfidence,
+      oldBrainId: item.currentBrainId,
+      newBrainId,
+      brainChanged,
+      reassigned,
+    };
+  } catch (err) {
+    return {
+      itemId: item.id,
+      oldConfidenceScore: item.confidenceScore,
+      newConfidenceScore: item.confidenceScore,
+      oldBrainId: item.currentBrainId,
+      newBrainId: item.currentBrainId,
+      brainChanged: false,
+      reassigned: false,
+      error: err instanceof Error ? err.message : String(err),
+    };
+  }
+}
+
+// ── Step 3: TriageQualityReportAgent ──────────────────────────────────────────
+
+function buildTriageReport(
+  runId: string,
+  confidenceThreshold: number,
+  dryRun: boolean,
+  totalFetched: number,
+  results: RetriageResult[]
+): TriageQualityReport {
+  const now = new Date().toISOString();
+  const improved = results.filter(
+    r => r.newConfidenceScore > r.oldConfidenceScore && !r.error
+  ).length;
+  const brainChanges = results.filter(r => r.brainChanged).length;
+  const reassigned = results.filter(r => r.reassigned).length;
+  const failed = results.filter(r => !!r.error).length;
+
+  const summary = dryRun
+    ? `DRY RUN: ${results.length} items below confidence threshold ${confidenceThreshold} found. No retriaging performed.`
+    : results.length === 0
+      ? `All ${totalFetched} memory items meet confidence threshold ${confidenceThreshold}. No action needed.`
+      : `Retriaged ${results.length - failed}/${results.length} items. ${improved} improved confidence, ${brainChanges} brain routing changes, ${reassigned} items reassigned. ${failed} failed.`;
+
+  return {
+    runId,
+    productId: 'mindlyst',
+    confidenceThreshold,
+    dryRun,
+    totalFetched,
+    lowConfidenceCount: results.length,
+    retriaged: results.length - failed,
+    improved,
+    brainChanges,
+    reassigned,
+    failed,
+    perItem: results,
+    summary,
+    generatedAt: now,
+  };
+}
+
+// ── Pipeline runner ────────────────────────────────────────────────────────────
+
+async function runTriageQualityPipeline(
+  confidenceThreshold: number,
+  dryRun: boolean,
+  req: McpToolRequest
+): Promise<TriageQualityReport> {
+  const runId = randomUUID();
+  const opts = { token: req.headers.authorization?.slice(7), requestId: req.id };
+
+  req.log.info(
+    { runId, stepId: 'collect', confidenceThreshold, dryRun },
+    'LowConfidenceCollectorAgent start'
+  );
+  const { items, totalFetched } = await collectLowConfidenceItems(confidenceThreshold, opts);
+  req.log.info(
+    { runId, stepId: 'collect', totalFetched, lowConfidenceCount: items.length },
+    'LowConfidenceCollectorAgent done'
+  );
+
+  req.log.info({ runId, stepId: 'retriage', count: items.length, dryRun }, 'RetriageAgent start');
+  const results: RetriageResult[] = [];
+  for (const item of items) {
+    const result = await retriageItem(item, dryRun, opts);
+    results.push(result);
+  }
+  req.log.info(
+    {
+      runId,
+      stepId: 'retriage',
+      improved: results.filter(r => r.newConfidenceScore > r.oldConfidenceScore).length,
+    },
+    'RetriageAgent done'
+  );
+
+  req.log.info({ runId, stepId: 'report' }, 'TriageQualityReportAgent start');
+  const report = buildTriageReport(runId, confidenceThreshold, dryRun, totalFetched, results);
+  req.log.info(
+    { runId, stepId: 'report', summary: report.summary },
+    'TriageQualityReportAgent done'
+  );
+
+  return report;
+}
+
+// ── MCP tool registration ─────────────────────────────────────────────────────
+
+registerTool({
+  name: 'mindlyst.memory.triageQuality',
+  description:
+    'A2A pipeline: finds MindLyst memory items with confidenceScore below a threshold, re-runs extraction on each, and auto-reassigns items where brain routing has changed. Returns a quality improvement report with confidence deltas and brain reassignment counts. Use dryRun=true to preview. Requires admin role.',
+  requiredRole: 'admin',
+  inputSchema: z.object({
+    confidenceThreshold: z.coerce
+      .number()
+      .min(0)
+      .max(1)
+      .default(0.5)
+      .describe('Items with confidenceScore below this are re-triaged (default 0.5)'),
+    dryRun: z
+      .boolean()
+      .default(false)
+      .describe('If true, collect and count only — do not retriage or reassign'),
+  }),
+  async execute(args, req) {
+    return runTriageQualityPipeline(args.confidenceThreshold, args.dryRun, req);
+  },
+});
diff --git a/services/mcp-server/src/server.ts b/services/mcp-server/src/server.ts
index 92f83266..cc43ed70 100644
--- a/services/mcp-server/src/server.ts
+++ b/services/mcp-server/src/server.ts
@@ -42,6 +42,9 @@ import './modules/a2a/sync-diagnostics-pipeline.js';
 import './modules/a2a/transcript-extraction-pipeline.js';
 import './modules/a2a/sync-conflict-pipeline.js';
 import './modules/a2a/route-safety-pipeline.js';
+import './modules/a2a/memory-curation-pipeline.js';
+import './modules/a2a/engagement-pipeline.js';
+import './modules/a2a/triage-quality-pipeline.js';
 import './modules/mindlyst/mindlyst-tools.js';
 import './modules/lysnrai/lysnrai-tools.js';
 import './modules/jarvis/jarvis-tools.js';