AgentWorkforce · khaliqgant · Dec 26, 2025 · Dec 26, 2025 · Dec 26, 2025 · Dec 26, 2025
@@ -29,6 +29,7 @@
 {"id":"agent-relay-5fa","title":"Add exponential backoff for daemon reconnection","description":"Implement graceful reconnection with exponential backoff delays [100, 500, 1000, 2000, 5000ms]. After max attempts, operate offline gracefully. See docs/TMUX_IMPROVEMENTS.md for implementation details.","status":"closed","priority":2,"issue_type":"feature","assignee":"LeadDev","created_at":"2025-12-20T21:28:48.055013+01:00","updated_at":"2025-12-20T21:33:42.229756+01:00","closed_at":"2025-12-20T21:33:42.229756+01:00"}
 {"id":"agent-relay-5g0","title":"Heartbeat timeout could be more configurable","description":"In connection.ts:196, heartbeat timeout is hardcoded as 2x heartbeatMs. This should be independently configurable. Also, heartbeat failures immediately kill the connection - could implement exponential backoff for transient issues.","status":"closed","priority":2,"issue_type":"task","created_at":"2025-12-20T00:18:03.556614+01:00","updated_at":"2025-12-23T23:03:07.563273+01:00","closed_at":"2025-12-23T23:03:07.563273+01:00"}
 {"id":"agent-relay-68a","title":"Ask pattern: request/response messaging (replyTo + deferred)","description":"Implement an RPC-like ask() over agent-relay: sender creates a short-lived deferred handle, sends message with replyTo, receiver responds resolving the deferred. Similar to swarm-mail DurableDeferred + ask/respond.","acceptance_criteria":"- ask() returns response or times out\\n- Deferreds are persisted with TTL\\n- Works across process restarts/reconnects","status":"open","priority":2,"issue_type":"task","created_at":"2025-12-20T21:44:33.930531+01:00","updated_at":"2025-12-20T21:44:33.930531+01:00","labels":["coordination","protocol"]}
+{"id":"agent-relay-6dl8","title":"Multi-line relay messages truncated in dashboard display","description":"Multi-line messages sent via -\u003erelay: are being cut off after first line when displayed in dashboard. This blocks communication between agents. Need to: (1) Verify daemon sends full message, (2) Fix dashboard rendering to handle newlines properly.","status":"closed","priority":1,"issue_type":"bug","created_at":"2025-12-26T12:28:39.746942+01:00","updated_at":"2025-12-26T12:38:36.844034+01:00","closed_at":"2025-12-26T12:38:36.844034+01:00"}
 {"id":"agent-relay-6mo","title":"Lead spawn capability not working (@relay:spawn pattern)","description":"The lead command advertises spawn capability via @relay:spawn pattern but it's not fully implemented. Agents started with 'agent-relay lead' should be able to spawn workers, but the parser extension mentioned in the code is not implemented.","status":"closed","priority":1,"issue_type":"bug","created_at":"2025-12-23T12:14:59.492237+01:00","updated_at":"2025-12-23T12:18:14.114911+01:00","closed_at":"2025-12-23T12:18:14.114911+01:00"}
 {"id":"agent-relay-6nx","title":"Add activity state tracking for better injection timing","description":"Track active/idle/disconnected state with timestamps. When session goes idle (30s no activity), trigger message injection opportunity. Improves injection timing vs current fixed 1.5s wait. See docs/TMUX_IMPROVEMENTS.md for implementation details.","status":"closed","priority":2,"issue_type":"feature","assignee":"LeadDev","created_at":"2025-12-20T21:28:46.993856+01:00","updated_at":"2025-12-20T21:33:42.223042+01:00","closed_at":"2025-12-20T21:33:42.223042+01:00"}
 {"id":"agent-relay-6ny","title":"Fix message truncation: store full message in SQLite first, include ID in relay","description":"","status":"closed","priority":1,"issue_type":"bug","created_at":"2025-12-19T22:04:36.168862+01:00","updated_at":"2025-12-19T22:08:28.207532+01:00","closed_at":"2025-12-19T22:08:28.207532+01:00"}
@@ -78,7 +79,7 @@
 {"id":"agent-relay-fb3","title":"Add configurable relay prefix for Gemini compatibility","description":"Gemini CLI uses @ for file references, conflicting with @relay:. Add --prefix flag to CLI and relayPrefix config option. Auto-detect CLI type and use appropriate default (\u003e\u003e for Gemini, @relay: for Claude/Codex). Update parser to use dynamic prefix. See docs/TMUX_IMPROVEMENTS.md for full implementation plan.","status":"closed","priority":2,"issue_type":"feature","assignee":"Coordinator","created_at":"2025-12-20T21:28:52.685002+01:00","updated_at":"2025-12-20T21:40:13.066766+01:00","closed_at":"2025-12-20T21:40:13.066766+01:00"}
 {"id":"agent-relay-ghy","title":"Team config: auto-spawn agents or auto-assign names from teams.json","description":"When a project has a teams.json config file, agent-relay up should either:\n\n1. **Auto-spawn option**: Automatically kick off terminal sessions for each agent defined in teams.json\n2. **Auto-assign option**: When users manually start sessions with `agent-relay -n \u003cname\u003e claude`, validate the name against teams.json and auto-assign roles/permissions\n\n## teams.json format\n```json\n{\n  \"team\": \"my-project\",\n  \"agents\": [\n    {\"name\": \"Coordinator\", \"cli\": \"claude\", \"role\": \"coordinator\"},\n    {\"name\": \"LeadDev\", \"cli\": \"claude\", \"role\": \"developer\"},\n    {\"name\": \"Reviewer\", \"cli\": \"claude\", \"role\": \"reviewer\"}\n  ],\n  \"autoSpawn\": false\n}\n```\n\n## Behavior\n- If `autoSpawn: true`, `agent-relay up` spawns tmux sessions for each agent\n- If `autoSpawn: false`, validate names against config when agents connect\n- Store teams.json in project root or .agent-relay/teams.json\n\n## Commands\n- `agent-relay up --spawn` - force spawn all agents\n- `agent-relay up --no-spawn` - just start daemon, manual agent starts","status":"open","priority":2,"issue_type":"feature","created_at":"2025-12-19T23:13:02.482971+01:00","updated_at":"2025-12-22T22:11:38.639588+01:00"}
 {"id":"agent-relay-go9","title":"PostgreSQL storage adapter not implemented","description":"In storage/adapter.ts:152-162, PostgreSQL is listed as a storage option but throws 'not yet implemented'. For production multi-node deployments, SQLite won't scale. Implement PostgreSQL adapter for distributed storage.","status":"open","priority":2,"issue_type":"feature","created_at":"2025-12-20T00:17:45.065487+01:00","updated_at":"2025-12-20T00:17:45.065487+01:00"}
-{"id":"agent-relay-hgd","title":"Reliable delivery: resume/replay via cursor/checkpoint","description":"Make ACK/RESUME real: persist a per-agent delivery cursor (last delivered/acked seq) and on reconnect replay missed messages from storage. Use existing delivery.seq as stream position, similar to swarm-mail DurableCursor checkpointing.","acceptance_criteria":"- Client can reconnect and receive missed messages\\n- Duplicate suppression is deterministic (id/seq based)\\n- Cursor persists across daemon restarts","status":"in_progress","priority":1,"issue_type":"task","assignee":"BE-Dev","created_at":"2025-12-20T21:44:02.016428+01:00","updated_at":"2025-12-24T14:29:48.822052+01:00","labels":["durability","protocol"]}
+{"id":"agent-relay-hgd","title":"Reliable delivery: resume/replay via cursor/checkpoint","description":"Make ACK/RESUME real: persist a per-agent delivery cursor (last delivered/acked seq) and on reconnect replay missed messages from storage. Use existing delivery.seq as stream position, similar to swarm-mail DurableCursor checkpointing.","acceptance_criteria":"- Client can reconnect and receive missed messages\\n- Duplicate suppression is deterministic (id/seq based)\\n- Cursor persists across daemon restarts","notes":"Persistent resume tokens with stored cursors; replay unacked messages on reconnect; ack tracking + dedup; tests passing.","status":"in_progress","priority":1,"issue_type":"task","assignee":"BE-Dev","created_at":"2025-12-20T21:44:02.016428+01:00","updated_at":"2025-12-26T12:13:54.547885+01:00","labels":["durability","protocol"]}
 {"id":"agent-relay-hgn","title":"Agent team hierarchies with specialized sub-agents","description":"Each main agent should be able to spawn and coordinate a team of specialized sub-agents:\n\n- **Lead** → Operations Consultant (monitors behavior, provides retrospective feedback on coordination)\n- **Implementer** → Code Reviewer (reviews code before commits, suggests improvements)\n- **Designer** → UX Reviewer (validates design decisions)\n- **Architect** → Security Auditor (checks for vulnerabilities)\n\nImplementation ideas:\n1. Define agent profiles with 'team' configurations in .claude/agents/\n2. When an agent starts, it can auto-spawn its team members\n3. Sub-agents monitor the parent agent's work and provide feedback\n4. At session end, sub-agents provide retrospective summaries\n5. Use -\u003erelay:spawn to create sub-agents with specific roles\n\nThis enables each agent to have a dedicated support team for quality assurance and improvement.","status":"open","priority":2,"issue_type":"feature","created_at":"2025-12-23T15:03:56.608847+01:00","updated_at":"2025-12-23T15:03:56.608847+01:00"}
 {"id":"agent-relay-hgw","title":"PR-9 Review: Add reconnection logic to MultiProjectClient","description":"","status":"closed","priority":2,"issue_type":"task","created_at":"2025-12-22T21:54:09.712003+01:00","updated_at":"2025-12-24T11:50:55.708361+01:00","closed_at":"2025-12-24T11:50:55.708361+01:00"}
 {"id":"agent-relay-hks","title":"Increase test coverage for daemon/server.ts, dashboard, and CLI","description":"Current coverage is only 39% overall. Key files with 0% coverage: daemon/server.ts, dashboard/server.ts, cli/index.ts, wrapper/client.ts, wrapper/tmux-wrapper.ts, utils/project-namespace.ts. Need integration tests for the daemon startup/shutdown lifecycle and CLI commands.","status":"closed","priority":2,"issue_type":"task","assignee":"Lead","created_at":"2025-12-20T00:17:29.603137+01:00","updated_at":"2025-12-22T17:14:41.611717+01:00","closed_at":"2025-12-22T17:14:41.611717+01:00"}
@@ -130,7 +131,7 @@
 {"id":"agent-relay-ytt","title":"Implement threaded conversations (Slack-style)","description":"Add thread replies with parent_message_id tracking in storage, nested thread UI display in dashboard, thread notification badges, and expand/collapse thread views. Backend: add parent_message_id column, thread reply routing. Frontend: inline thread expansion, reply count badges, thread-specific compose.","status":"closed","priority":1,"issue_type":"feature","created_at":"2025-12-23T12:00:00Z","updated_at":"2025-12-23T16:37:22.987643+01:00","closed_at":"2025-12-23T16:37:22.987643+01:00"}
 {"id":"agent-relay-ytu","title":"[Control Plane] Design Control API (REST + WebSocket)","description":"Design API for human control plane: POST /tasks, GET /agents, POST /agents/:id/msg, WS /stream. OpenAPI spec. Authentication endpoints. See ai-maestro's manager/worker pattern for reference.","status":"open","priority":2,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-21T14:00:00Z"}
 {"id":"agent-relay-ytv","title":"[Control Plane] Implement Lead Agent orchestration","description":"Lead Agent receives human intent, breaks into tasks, assigns to specialized agents based on skills, monitors progress, escalates decisions. Can be AI-powered or rule-based initially.","status":"open","priority":2,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-21T14:00:00Z"}
-{"id":"agent-relay-ytw","title":"[Control Plane] Build web dashboard v2 (fleet control)","description":"Next.js dashboard with: fleet overview, task assignment UI, agent status cards, trajectory viewer, decision queue. Learn from ai-maestro's hierarchical naming and color coding.","status":"open","priority":2,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-21T14:00:00Z"}
+{"id":"agent-relay-ytw","title":"[Control Plane] Build web dashboard v2 (fleet control)","description":"Next.js dashboard with: fleet overview, task assignment UI, agent status cards, trajectory viewer, decision queue. Learn from ai-maestro's hierarchical naming and color coding.","status":"in_progress","priority":2,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-26T12:39:19.789712+01:00"}
 {"id":"agent-relay-ytx","title":"[Control Plane] Human authentication (OAuth/magic link)","description":"Secure human access to control plane. OAuth2 (GitHub/Google) or magic link email auth. JWT tokens for API access. Role-based permissions (admin/operator/viewer).","status":"open","priority":2,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-21T14:00:00Z"}
 {"id":"agent-relay-yty","title":"[Control Plane] Push notification service","description":"Real-time notifications to mobile/web: APNs for iOS, FCM for Android/web. Notify on: task completion, agent questions, escalations, errors.","status":"open","priority":3,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-21T14:00:00Z"}
 {"id":"agent-relay-ytz","title":"[Control Plane] iPhone app MVP","description":"React Native or Swift app: fleet status view, task list, agent messaging, push notifications, decision queue with quick actions. Minimal viable version.","status":"open","priority":3,"issue_type":"task","created_at":"2025-12-21T14:00:00Z","updated_at":"2025-12-21T14:00:00Z"}

@@ -1,8 +1,8 @@
 import { describe, it, expect, vi } from 'vitest';
 import type { Socket } from 'node:net';
 import { Connection } from './connection.js';
-import { encodeFrame } from '../protocol/framing.js';
-import { PROTOCOL_VERSION, type Envelope, type HelloPayload } from '../protocol/types.js';
+import { encodeFrame, FrameParser } from '../protocol/framing.js';
+import { PROTOCOL_VERSION, type Envelope, type HelloPayload, type WelcomePayload } from '../protocol/types.js';
 
 class MockSocket {
   private handlers: Map<string, Array<(...args: any[]) => void>> = new Map();
@@ -82,6 +82,36 @@ describe('Connection', () => {
     expect(socket.destroyed).toBe(true);
   });
 
+  it('accepts resume token when resumeHandler returns a session', async () => {
+    const socket = new MockSocket();
+    const parser = new FrameParser();
+    const resumeHandler = vi.fn().mockResolvedValue({
+      sessionId: 'session-resume',
+      resumeToken: 'token-abc',
+      seedSequences: [{ topic: 'chat', peer: 'peerA', seq: 5 }],
+    });
+    const connection = new Connection(socket as unknown as Socket, { heartbeatMs: 50, resumeHandler });
+    const onActive = vi.fn();
+    connection.onActive = onActive;
+
+    const hello = makeHello('agent-a');
+    hello.payload.session = { resume_token: 'token-abc' };
+    socket.emit('data', encodeFrame(hello));
+
+    await new Promise((r) => setTimeout(r, 0));
+
+    expect(connection.state).toBe('ACTIVE');
+    expect(connection.sessionId).toBe('session-resume');
+    expect(connection.resumeToken).toBe('token-abc');
+    expect(connection.isResumed).toBe(true);
+    expect(connection.getNextSeq('chat', 'peerA')).toBe(6); // seeded to 5, next is 6
+    expect(onActive).toHaveBeenCalledTimes(1);
+
+    const welcome = parser.push(socket.written[0])?.[0] as Envelope<WelcomePayload>;
+    expect(welcome.payload.resume_token).toBe('token-abc');
+    expect(welcome.payload.session_id).toBe('session-resume');
+  });
+
   describe('heartbeat timeout configuration', () => {
     it('uses configurable heartbeatTimeoutMultiplier', async () => {
       const socket = new MockSocket();

@@ -29,6 +29,12 @@ export interface ConnectionConfig {
   heartbeatMs: number;
   /** Multiplier for heartbeat timeout (timeout = heartbeatMs * multiplier). Default: 6 (30s with 5s heartbeat) */
   heartbeatTimeoutMultiplier: number;
+  /** Optional handler to validate resume tokens and provide session state */
+  resumeHandler?: (params: { agent: string; resumeToken: string }) => Promise<{
+    sessionId: string;
+    resumeToken?: string;
+    seedSequences?: Array<{ topic?: string; peer: string; seq: number }>;
+  } | null>;
 }
 
 export const DEFAULT_CONFIG: ConnectionConfig = {
@@ -53,6 +59,7 @@ export class Connection {
   private _workingDirectory?: string;
   private _sessionId: string;
   private _resumeToken: string;
+  private _isResumed = false;
 
   private heartbeatTimer?: NodeJS.Timeout;
   private lastPongReceived?: number;
@@ -112,6 +119,14 @@ export class Connection {
     return this._sessionId;
   }
 
+  get resumeToken(): string {
+    return this._resumeToken;
+  }
+
+  get isResumed(): boolean {
+    return this._isResumed;
+  }
+
   private setupSocketHandlers(): void {
     this.socket.on('data', (data) => this.handleData(data));
     this.socket.on('close', () => this.handleClose());
@@ -122,18 +137,21 @@ export class Connection {
     try {
       const frames = this.parser.push(data);
       for (const frame of frames) {
-        this.processFrame(frame);
+        this.processFrame(frame).catch((err) => {
+          this.sendError('BAD_REQUEST', `Frame error: ${err}`, true);
+          this.close();
+        });
       }
     } catch (err) {
       this.sendError('BAD_REQUEST', `Frame error: ${err}`, true);
       this.close();
     }
   }
 
-  private processFrame(envelope: Envelope): void {
+  private async processFrame(envelope: Envelope): Promise<void> {
     switch (envelope.type) {
       case 'HELLO':
-        this.handleHello(envelope as Envelope<HelloPayload>);
+        await this.handleHello(envelope as Envelope<HelloPayload>);
         break;
       case 'SEND':
         this.handleSend(envelope as Envelope<SendPayload>);
@@ -156,7 +174,7 @@ export class Connection {
     }
   }
 
-  private handleHello(envelope: Envelope<HelloPayload>): void {
+  private async handleHello(envelope: Envelope<HelloPayload>): Promise<void> {
     if (this._state !== 'HANDSHAKING') {
       this.sendError('BAD_REQUEST', 'Unexpected HELLO', false);
       return;
@@ -170,22 +188,44 @@ export class Connection {
     this._workingDirectory = envelope.payload.workingDirectory;
 
     // Check for session resume
-    if (envelope.payload.session?.resume_token) {
-      // Resume tokens are not persisted; tell client to start a fresh session.
-      this.sendError('RESUME_TOO_OLD', 'Session resume not yet supported; starting new session', false);
+    const resumeToken = envelope.payload.session?.resume_token;
+    if (resumeToken) {
+      if (this.config.resumeHandler) {
+        try {
+          const resumeState = await this.config.resumeHandler({
+            agent: this._agentName,
+            resumeToken,
+          });
+
+          if (resumeState) {
+            this._sessionId = resumeState.sessionId;
+            this._resumeToken = resumeState.resumeToken ?? resumeToken;
+            this._isResumed = true;
+
+            // Seed sequence counters so new deliveries continue from last known seq per stream
+            for (const seed of resumeState.seedSequences ?? []) {
+              this.seedSequence(seed.topic ?? 'default', seed.peer, seed.seq);
+            }
+          } else {
+            this.sendError('RESUME_TOO_OLD', 'Resume token rejected; starting new session', false);
+          }
+        } catch (err: any) {
+          this.sendError('RESUME_TOO_OLD', `Resume validation failed: ${err?.message ?? err}`, false);
+        }
+      } else {
+        this.sendError('RESUME_TOO_OLD', 'Session resume not configured; starting new session', false);
+      }
     }
 
     // Send WELCOME
-    // Note: resume_token is omitted because session resume is not yet implemented.
-    // Sending a token would cause clients to attempt resume on reconnect,
-    // triggering a RESUME_TOO_OLD -> new token -> reconnect loop.
     const welcome: Envelope<WelcomePayload> = {
       v: PROTOCOL_VERSION,
       type: 'WELCOME',
       id: uuid(),
       ts: Date.now(),
       payload: {
         session_id: this._sessionId,
+        resume_token: this._resumeToken,
         server: {
           max_frame_bytes: this.config.maxFrameBytes,
           heartbeat_ms: this.config.heartbeatMs,
@@ -256,6 +296,17 @@ export class Connection {
     }
   }
 
+  /**
+   * Seed a sequence counter for a stream so that the next value continues from the provided seq.
+   */
+  seedSequence(topic: string | undefined, peer: string, seq: number): void {
+    const key = `${topic ?? 'default'}:${peer}`;
+    const current = this.sequences.get(key) ?? 0;
+    if (seq > current) {
+      this.sequences.set(key, seq);
+    }
+  }
+
   /**
    * Get next sequence number for a stream.
    */