@ruvector/edge-net 0.1.2 → 0.1.3

package/README.md

@@ -44,6 +44,7 @@ A distributed computing platform that enables collective resource sharing for AI
 - [Exotic AI Capabilities](#exotic-ai-capabilities)
 - [Core Architecture & Capabilities](#core-architecture--capabilities)
 - [Self-Learning Hooks & MCP Integration](#self-learning-hooks--mcp-integration)
+- [Distributed AI Agents & Workers](#distributed-ai-agents--workers)
 
 ---
 
@@ -1238,6 +1239,165 @@ ruvector hooks remember <content> -t <type>  # Store memory
 ruvector hooks recall <query>                # Semantic search
 ```
 
+---
+
+## Distributed AI Agents & Workers
+
+Edge-net enables spawning AI agents and distributed worker pools across the collective compute network. This transforms passive compute contribution into active distributed AI execution.
+
+### Architecture
+
+```
+┌─────────────────────────────────────────────────────────────────────────────┐
+│                    DISTRIBUTED AI AGENT SYSTEM                               │
+├─────────────────────────────────────────────────────────────────────────────┤
+│                                                                             │
+│  ┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐       │
+│  │    AgentSpawner │     │   WorkerPool    │     │ TaskOrchestrator│       │
+│  │                 │     │                 │     │                 │       │
+│  │ • Type routing  │     │ • Load balance  │     │ • Workflows     │       │
+│  │ • rUv costing   │     │ • Auto-scaling  │     │ • Dependencies  │       │
+│  │ • Priority mgmt │     │ • Fault tolerant│     │ • Parallel exec │       │
+│  └────────┬────────┘     └────────┬────────┘     └────────┬────────┘       │
+│           │                       │                       │                 │
+│           └───────────────────────┼───────────────────────┘                 │
+│                                   │                                         │
+│                      ┌────────────┴────────────┐                            │
+│                      │   Edge-Net P2P Network   │                            │
+│                      │   (WebRTC Data Channels) │                            │
+│                      └──────────────────────────┘                            │
+│                                                                             │
+│   Agent Types:  researcher | coder | reviewer | tester | analyst |          │
+│                 optimizer | coordinator | embedder                          │
+│                                                                             │
+└─────────────────────────────────────────────────────────────────────────────┘
+```
+
+### Agent Types
+
+| Type | Capabilities | Base rUv | Use Cases |
+|------|-------------|----------|-----------|
+| **researcher** | search, analyze, summarize, extract | 10 | Codebase analysis, documentation research |
+| **coder** | code, refactor, debug, test | 15 | Feature implementation, bug fixes |
+| **reviewer** | review, audit, validate, suggest | 12 | Code review, security audit |
+| **tester** | test, benchmark, validate, report | 10 | Test creation, coverage analysis |
+| **analyst** | analyze, metrics, report, visualize | 8 | Performance analysis, data insights |
+| **optimizer** | optimize, profile, benchmark, improve | 15 | Performance tuning, efficiency |
+| **coordinator** | orchestrate, route, schedule, monitor | 20 | Multi-agent workflows |
+| **embedder** | embed, vectorize, similarity, search | 5 | Vector operations, semantic search |
+
+### CLI Commands
+
+```bash
+# Show Edge-Net information
+ruvector edge-net info
+
+# Spawn a distributed AI agent
+ruvector edge-net spawn researcher "Analyze the authentication system"
+ruvector edge-net spawn coder "Implement user profile feature" --max-ruv 50 --priority high
+
+# Create and use worker pools
+ruvector edge-net pool create --size 10 --capabilities compute,embed
+ruvector edge-net pool execute "Process batch embeddings"
+
+# Run multi-agent workflows
+ruvector edge-net workflow code-review
+ruvector edge-net workflow feature-dev
+ruvector edge-net workflow optimization
+
+# Check network status
+ruvector edge-net status
+```
+
+### MCP Tools
+
+The following MCP tools are available when `ruvector` is configured as an MCP server:
+
+| Tool | Description | Parameters |
+|------|-------------|------------|
+| `edge_net_info` | Get Edge-Net information | - |
+| `edge_net_spawn` | Spawn distributed agent | type, task, max_ruv, priority |
+| `edge_net_pool_create` | Create worker pool | min_workers, max_workers |
+| `edge_net_pool_execute` | Execute on pool | task, pool_id |
+| `edge_net_workflow` | Run workflow | name (code-review, feature-dev, etc.) |
+| `edge_net_status` | Network status | - |
+
+### Workflows
+
+Pre-built multi-agent workflows:
+
+| Workflow | Steps | Est. rUv | Description |
+|----------|-------|----------|-------------|
+| **code-review** | analyst → reviewer → tester → optimizer | 45 | Comprehensive code analysis |
+| **feature-dev** | researcher → coder → tester → reviewer | 60 | Full feature development cycle |
+| **bug-fix** | analyst → coder → tester | 35 | Bug diagnosis and fix |
+| **optimization** | analyst → optimizer → coder → tester | 50 | Performance improvement |
+| **research** | researcher → analyst → embedder | 30 | Deep research with embeddings |
+
+### JavaScript API
+
+```javascript
+import { AgentSpawner, WorkerPool, TaskOrchestrator, AGENT_TYPES } from '@ruvector/edge-net/agents';
+
+// Spawn a distributed agent
+const spawner = new AgentSpawner(edgeNetNode);
+const agent = await spawner.spawn('coder', 'Implement authentication', {
+    maxRuv: 30,
+    priority: 'high'
+});
+
+// Create a worker pool
+const pool = new WorkerPool(edgeNetNode, {
+    minWorkers: 5,
+    maxWorkers: 20,
+    capabilities: ['compute', 'embed', 'analyze']
+});
+await pool.scale(10);
+
+// Execute tasks on the pool
+const result = await pool.execute({
+    type: 'parallel',
+    task: 'Process batch data',
+    data: largeDataset
+});
+
+// Run multi-agent workflow
+const orchestrator = new TaskOrchestrator(edgeNetNode, spawner);
+await orchestrator.runWorkflow('feature-dev', 'Add user authentication');
+```
+
+### Event System
+
+Agents and workers emit events for monitoring:
+
+```javascript
+agent.on('started', ({ id, type }) => console.log(`Agent ${id} started`));
+agent.on('progress', ({ progress }) => console.log(`Progress: ${progress}%`));
+agent.on('completed', ({ result }) => console.log('Done:', result));
+agent.on('error', ({ error }) => console.error('Error:', error));
+
+pool.on('scaled', ({ workers }) => console.log(`Pool scaled to ${workers}`));
+pool.on('task_completed', ({ taskId }) => console.log(`Task ${taskId} done`));
+```
+
+### rUv Economics for Agents
+
+| Factor | Impact |
+|--------|--------|
+| **Base Cost** | Agent type determines base rUv per task |
+| **Task Complexity** | Longer/complex tasks cost more |
+| **Priority** | High priority = 1.5x cost, Critical = 2x |
+| **Network Load** | Dynamic pricing based on availability |
+| **Early Adopter** | 10x multiplier during genesis phase |
+
+### Security Considerations
+
+- All agent communications are encrypted via DTLS
+- Task execution sandboxed in WebWorkers
+- rUv spending limits prevent runaway costs
+- Input validation on all MCP tools
+- Rate limiting on agent spawning
+
 ### Claude Code Hook Events
 
 | Event | Trigger | Action |

package/agents.js

@@ -0,0 +1,965 @@
+#!/usr/bin/env node
+/**
+ * Edge-Net Agent System
+ *
+ * Distributed AI agent execution across the Edge-Net collective.
+ * Spawn agents, create worker pools, and orchestrate multi-agent workflows.
+ */
+
+import { EventEmitter } from 'events';
+import { randomBytes, createHash } from 'crypto';
+
+// Agent types and their capabilities
+export const AGENT_TYPES = {
+    researcher: {
+        name: 'Researcher',
+        capabilities: ['search', 'analyze', 'summarize', 'extract'],
+        baseRuv: 10,
+        description: 'Analyzes and researches information',
+    },
+    coder: {
+        name: 'Coder',
+        capabilities: ['code', 'refactor', 'debug', 'test'],
+        baseRuv: 15,
+        description: 'Writes and improves code',
+    },
+    reviewer: {
+        name: 'Reviewer',
+        capabilities: ['review', 'audit', 'validate', 'suggest'],
+        baseRuv: 12,
+        description: 'Reviews code and provides feedback',
+    },
+    tester: {
+        name: 'Tester',
+        capabilities: ['test', 'benchmark', 'validate', 'report'],
+        baseRuv: 10,
+        description: 'Tests and validates implementations',
+    },
+    analyst: {
+        name: 'Analyst',
+        capabilities: ['analyze', 'metrics', 'report', 'visualize'],
+        baseRuv: 8,
+        description: 'Analyzes data and generates reports',
+    },
+    optimizer: {
+        name: 'Optimizer',
+        capabilities: ['optimize', 'profile', 'benchmark', 'improve'],
+        baseRuv: 15,
+        description: 'Optimizes performance and efficiency',
+    },
+    coordinator: {
+        name: 'Coordinator',
+        capabilities: ['orchestrate', 'route', 'schedule', 'monitor'],
+        baseRuv: 20,
+        description: 'Coordinates multi-agent workflows',
+    },
+    embedder: {
+        name: 'Embedder',
+        capabilities: ['embed', 'vectorize', 'similarity', 'search'],
+        baseRuv: 5,
+        description: 'Generates embeddings and vector operations',
+    },
+};
+
+// Task status enum
+export const TaskStatus = {
+    PENDING: 'pending',
+    QUEUED: 'queued',
+    ASSIGNED: 'assigned',
+    RUNNING: 'running',
+    COMPLETED: 'completed',
+    FAILED: 'failed',
+    CANCELLED: 'cancelled',
+};
+
+/**
+ * Distributed Agent
+ *
+ * Represents an AI agent running on the Edge-Net network.
+ */
+export class DistributedAgent extends EventEmitter {
+    constructor(options) {
+        super();
+        this.id = `agent-${randomBytes(8).toString('hex')}`;
+        this.type = options.type || 'researcher';
+        this.task = options.task;
+        this.config = AGENT_TYPES[this.type] || AGENT_TYPES.researcher;
+        this.maxRuv = options.maxRuv || this.config.baseRuv;
+        this.priority = options.priority || 'medium';
+        this.timeout = options.timeout || 300000; // 5 min default
+
+        this.status = TaskStatus.PENDING;
+        this.assignedNode = null;
+        this.progress = 0;
+        this.result = null;
+        this.error = null;
+        this.startTime = null;
+        this.endTime = null;
+        this.ruvSpent = 0;
+
+        this.subtasks = [];
+        this.logs = [];
+    }
+
+    /**
+     * Get agent info
+     */
+    getInfo() {
+        return {
+            id: this.id,
+            type: this.type,
+            task: this.task,
+            status: this.status,
+            progress: this.progress,
+            assignedNode: this.assignedNode,
+            maxRuv: this.maxRuv,
+            ruvSpent: this.ruvSpent,
+            startTime: this.startTime,
+            endTime: this.endTime,
+            duration: this.endTime && this.startTime
+                ? this.endTime - this.startTime
+                : null,
+        };
+    }
+
+    /**
+     * Update agent progress
+     */
+    updateProgress(progress, message) {
+        this.progress = Math.min(100, Math.max(0, progress));
+        this.log(`Progress: ${this.progress}% - ${message}`);
+        this.emit('progress', { progress: this.progress, message });
+    }
+
+    /**
+     * Log message
+     */
+    log(message) {
+        const entry = {
+            timestamp: Date.now(),
+            message,
+        };
+        this.logs.push(entry);
+        this.emit('log', entry);
+    }
+
+    /**
+     * Mark as completed
+     */
+    complete(result) {
+        this.status = TaskStatus.COMPLETED;
+        this.result = result;
+        this.progress = 100;
+        this.endTime = Date.now();
+        this.log('Agent completed successfully');
+        this.emit('complete', result);
+    }
+
+    /**
+     * Mark as failed
+     */
+    fail(error) {
+        this.status = TaskStatus.FAILED;
+        this.error = error;
+        this.endTime = Date.now();
+        this.log(`Agent failed: ${error}`);
+        this.emit('error', error);
+    }
+
+    /**
+     * Cancel the agent
+     */
+    cancel() {
+        this.status = TaskStatus.CANCELLED;
+        this.endTime = Date.now();
+        this.log('Agent cancelled');
+        this.emit('cancelled');
+    }
+}
+
+/**
+ * Agent Spawner
+ *
+ * Spawns and manages distributed agents across the Edge-Net network.
+ */
+export class AgentSpawner extends EventEmitter {
+    constructor(networkManager, options = {}) {
+        super();
+        this.network = networkManager;
+        this.agents = new Map();
+        this.pendingQueue = [];
+        this.maxConcurrent = options.maxConcurrent || 10;
+        this.defaultTimeout = options.defaultTimeout || 300000;
+
+        // Agent routing table (learned from outcomes)
+        this.routingTable = new Map();
+
+        // Stats
+        this.stats = {
+            totalSpawned: 0,
+            completed: 0,
+            failed: 0,
+            totalRuvSpent: 0,
+        };
+    }
+
+    /**
+     * Spawn a new agent on the network
+     */
+    async spawn(options) {
+        const agent = new DistributedAgent({
+            ...options,
+            timeout: options.timeout || this.defaultTimeout,
+        });
+
+        this.agents.set(agent.id, agent);
+        this.stats.totalSpawned++;
+
+        agent.log(`Agent spawned: ${agent.type} - ${agent.task}`);
+        agent.status = TaskStatus.QUEUED;
+
+        // Find best node for this agent type
+        const targetNode = await this.findBestNode(agent);
+
+        if (targetNode) {
+            await this.assignToNode(agent, targetNode);
+        } else {
+            // Queue for later assignment
+            this.pendingQueue.push(agent);
+            agent.log('Queued - waiting for available node');
+        }
+
+        this.emit('agent-spawned', agent);
+        return agent;
+    }
+
+    /**
+     * Find the best node for an agent based on capabilities and load
+     */
+    async findBestNode(agent) {
+        if (!this.network) return null;
+
+        const peers = this.network.getPeerList ?
+            this.network.getPeerList() :
+            Array.from(this.network.peers?.values() || []);
+
+        if (peers.length === 0) return null;
+
+        // Score each peer based on:
+        // 1. Capability match
+        // 2. Current load
+        // 3. Historical performance
+        // 4. Latency
+        const scoredPeers = peers.map(peer => {
+            let score = 50; // Base score
+
+            // Check capabilities
+            const peerCaps = peer.capabilities || [];
+            const requiredCaps = agent.config.capabilities;
+            const capMatch = requiredCaps.filter(c => peerCaps.includes(c)).length;
+            score += capMatch * 10;
+
+            // Check load (lower is better)
+            const load = peer.load || 0;
+            score -= load * 20;
+
+            // Check historical performance
+            const history = this.routingTable.get(`${peer.piKey || peer.id}-${agent.type}`);
+            if (history) {
+                score += history.successRate * 30;
+                score -= history.avgLatency / 1000; // Penalize high latency
+            }
+
+            return { peer, score };
+        });
+
+        // Sort by score (highest first)
+        scoredPeers.sort((a, b) => b.score - a.score);
+
+        return scoredPeers[0]?.peer || null;
+    }
+
+    /**
+     * Assign agent to a specific node
+     */
+    async assignToNode(agent, node) {
+        agent.status = TaskStatus.ASSIGNED;
+        agent.assignedNode = node.piKey || node.id;
+        agent.startTime = Date.now();
+        agent.log(`Assigned to node: ${agent.assignedNode.slice(0, 12)}...`);
+
+        // Send task to node via network
+        if (this.network?.sendToPeer) {
+            await this.network.sendToPeer(agent.assignedNode, {
+                type: 'agent_task',
+                agentId: agent.id,
+                agentType: agent.type,
+                task: agent.task,
+                maxRuv: agent.maxRuv,
+                timeout: agent.timeout,
+            });
+        }
+
+        agent.status = TaskStatus.RUNNING;
+        this.emit('agent-assigned', { agent, node });
+
+        // Set timeout
+        setTimeout(() => {
+            if (agent.status === TaskStatus.RUNNING) {
+                agent.fail('Timeout exceeded');
+            }
+        }, agent.timeout);
+    }
+
+    /**
+     * Handle task result from network
+     */
+    handleResult(agentId, result) {
+        const agent = this.agents.get(agentId);
+        if (!agent) return;
+
+        if (result.success) {
+            agent.complete(result.data);
+            this.stats.completed++;
+            this.updateRoutingTable(agent, true, result.latency);
+        } else {
+            agent.fail(result.error);
+            this.stats.failed++;
+            this.updateRoutingTable(agent, false, result.latency);
+        }
+
+        agent.ruvSpent = result.ruvSpent || agent.config.baseRuv;
+        this.stats.totalRuvSpent += agent.ruvSpent;
+    }
+
+    /**
+     * Update routing table with outcome
+     */
+    updateRoutingTable(agent, success, latency) {
+        const key = `${agent.assignedNode}-${agent.type}`;
+        const existing = this.routingTable.get(key) || {
+            attempts: 0,
+            successes: 0,
+            totalLatency: 0,
+        };
+
+        existing.attempts++;
+        if (success) existing.successes++;
+        existing.totalLatency += latency || 0;
+        existing.successRate = existing.successes / existing.attempts;
+        existing.avgLatency = existing.totalLatency / existing.attempts;
+
+        this.routingTable.set(key, existing);
+    }
+
+    /**
+     * Get agent by ID
+     */
+    getAgent(agentId) {
+        return this.agents.get(agentId);
+    }
+
+    /**
+     * List all agents
+     */
+    listAgents(filter = {}) {
+        let agents = Array.from(this.agents.values());
+
+        if (filter.status) {
+            agents = agents.filter(a => a.status === filter.status);
+        }
+        if (filter.type) {
+            agents = agents.filter(a => a.type === filter.type);
+        }
+
+        return agents.map(a => a.getInfo());
+    }
+
+    /**
+     * Get spawner stats
+     */
+    getStats() {
+        return {
+            ...this.stats,
+            activeAgents: Array.from(this.agents.values())
+                .filter(a => a.status === TaskStatus.RUNNING).length,
+            queuedAgents: this.pendingQueue.length,
+            successRate: this.stats.completed /
+                (this.stats.completed + this.stats.failed) || 0,
+        };
+    }
+}
+
+/**
+ * Worker Pool
+ *
+ * Manages a pool of distributed workers for parallel task execution.
+ */
+export class WorkerPool extends EventEmitter {
+    constructor(networkManager, options = {}) {
+        super();
+        this.id = `pool-${randomBytes(6).toString('hex')}`;
+        this.network = networkManager;
+        this.size = options.size || 5;
+        this.capabilities = options.capabilities || ['compute', 'embed'];
+        this.maxTasksPerWorker = options.maxTasksPerWorker || 10;
+
+        this.workers = new Map();
+        this.taskQueue = [];
+        this.activeTasks = new Map();
+        this.results = new Map();
+
+        this.status = 'initializing';
+        this.stats = {
+            tasksCompleted: 0,
+            tasksFailed: 0,
+            totalProcessingTime: 0,
+        };
+    }
+
+    /**
+     * Initialize the worker pool
+     */
+    async initialize() {
+        this.status = 'recruiting';
+        this.emit('status', 'Recruiting workers...');
+
+        // Find available workers from network
+        const peers = this.network?.getPeerList?.() ||
+            Array.from(this.network?.peers?.values() || []);
+
+        // Filter peers by capabilities
+        const eligiblePeers = peers.filter(peer => {
+            const peerCaps = peer.capabilities || [];
+            return this.capabilities.some(c => peerCaps.includes(c));
+        });
+
+        // Recruit up to pool size
+        const recruited = eligiblePeers.slice(0, this.size);
+
+        for (const peer of recruited) {
+            this.workers.set(peer.piKey || peer.id, {
+                id: peer.piKey || peer.id,
+                peer,
+                status: 'idle',
+                currentTasks: 0,
+                completedTasks: 0,
+                lastSeen: Date.now(),
+            });
+        }
+
+        // If not enough real workers, create virtual workers for local execution
+        while (this.workers.size < this.size) {
+            const virtualId = `virtual-${randomBytes(4).toString('hex')}`;
+            this.workers.set(virtualId, {
+                id: virtualId,
+                peer: null,
+                status: 'idle',
+                currentTasks: 0,
+                completedTasks: 0,
+                isVirtual: true,
+            });
+        }
+
+        this.status = 'ready';
+        this.emit('ready', {
+            poolId: this.id,
+            workers: this.workers.size,
+            realWorkers: Array.from(this.workers.values())
+                .filter(w => !w.isVirtual).length,
+        });
+
+        return this;
+    }
+
+    /**
+     * Execute tasks in parallel across workers
+     */
+    async execute(options) {
+        const {
+            task,
+            data,
+            strategy = 'parallel',
+            chunkSize = null,
+        } = options;
+
+        const batchId = `batch-${randomBytes(6).toString('hex')}`;
+        const startTime = Date.now();
+
+        // Split data into chunks for workers
+        let chunks;
+        if (Array.isArray(data)) {
+            const size = chunkSize || Math.ceil(data.length / this.workers.size);
+            chunks = [];
+            for (let i = 0; i < data.length; i += size) {
+                chunks.push(data.slice(i, i + size));
+            }
+        } else {
+            chunks = [data];
+        }
+
+        this.emit('batch-start', { batchId, chunks: chunks.length });
+
+        // Assign chunks to workers
+        const promises = chunks.map((chunk, index) =>
+            this.assignTask({
+                batchId,
+                index,
+                task,
+                data: chunk,
+            })
+        );
+
+        // Wait for all or handle based on strategy
+        let results;
+        if (strategy === 'parallel') {
+            results = await Promise.all(promises);
+        } else if (strategy === 'race') {
+            results = [await Promise.race(promises)];
+        } else {
+            // Sequential
+            results = [];
+            for (const promise of promises) {
+                results.push(await promise);
+            }
+        }
+
+        const endTime = Date.now();
+        this.stats.totalProcessingTime += endTime - startTime;
+
+        this.emit('batch-complete', {
+            batchId,
+            duration: endTime - startTime,
+            results: results.length,
+        });
+
+        // Flatten results if array
+        return Array.isArray(data) ? results.flat() : results[0];
+    }
+
+    /**
+     * Assign a single task to an available worker
+     */
+    async assignTask(taskInfo) {
+        const taskId = `task-${randomBytes(6).toString('hex')}`;
+
+        // Find idle worker
+        const worker = this.findIdleWorker();
+        if (!worker) {
+            // Queue task
+            return new Promise((resolve, reject) => {
+                this.taskQueue.push({ taskInfo, resolve, reject });
+            });
+        }
+
+        worker.status = 'busy';
+        worker.currentTasks++;
+
+        this.activeTasks.set(taskId, {
+            ...taskInfo,
+            workerId: worker.id,
+            startTime: Date.now(),
+        });
+
+        try {
+            // Execute on worker
+            const result = await this.executeOnWorker(worker, taskInfo);
+
+            worker.completedTasks++;
+            this.stats.tasksCompleted++;
+            this.results.set(taskId, result);
+
+            return result;
+        } catch (error) {
+            this.stats.tasksFailed++;
+            throw error;
+        } finally {
+            worker.currentTasks--;
+            if (worker.currentTasks === 0) {
+                worker.status = 'idle';
+            }
+            this.activeTasks.delete(taskId);
+
+            // Process queued task if any
+            if (this.taskQueue.length > 0) {
+                const { taskInfo, resolve, reject } = this.taskQueue.shift();
+                this.assignTask(taskInfo).then(resolve).catch(reject);
+            }
+        }
+    }
+
+    /**
+     * Find an idle worker
+     */
+    findIdleWorker() {
+        for (const worker of this.workers.values()) {
+            if (worker.status === 'idle' ||
+                worker.currentTasks < this.maxTasksPerWorker) {
+                return worker;
+            }
+        }
+        return null;
+    }
+
+    /**
+     * Execute task on a specific worker
+     */
+    async executeOnWorker(worker, taskInfo) {
+        if (worker.isVirtual) {
+            // Local execution for virtual workers
+            return this.executeLocally(taskInfo);
+        }
+
+        // Send to remote worker via network
+        return new Promise((resolve, reject) => {
+            const timeout = setTimeout(() => {
+                reject(new Error('Worker timeout'));
+            }, 60000);
+
+            // Send task
+            if (this.network?.sendToPeer) {
+                this.network.sendToPeer(worker.id, {
+                    type: 'worker_task',
+                    poolId: this.id,
+                    task: taskInfo.task,
+                    data: taskInfo.data,
+                });
+            }
+
+            // Listen for result
+            const handler = (msg) => {
+                if (msg.poolId === this.id && msg.batchId === taskInfo.batchId) {
+                    clearTimeout(timeout);
+                    this.network?.off?.('worker_result', handler);
+                    resolve(msg.result);
+                }
+            };
+
+            this.network?.on?.('worker_result', handler);
+
+            // Fallback to local if no response
+            setTimeout(() => {
+                clearTimeout(timeout);
+                this.executeLocally(taskInfo).then(resolve).catch(reject);
+            }, 5000);
+        });
+    }
+
+    /**
+     * Execute task locally (for virtual workers or fallback)
+     */
+    async executeLocally(taskInfo) {
+        const { task, data } = taskInfo;
+
+        // Simple local execution based on task type
+        switch (task) {
+            case 'embed':
+                // Simulate embedding
+                return Array.isArray(data)
+                    ? data.map(() => new Array(384).fill(0).map(() => Math.random()))
+                    : new Array(384).fill(0).map(() => Math.random());
+
+            case 'process':
+                return Array.isArray(data)
+                    ? data.map(item => ({ processed: true, item }))
+                    : { processed: true, data };
+
+            case 'analyze':
+                return {
+                    analyzed: true,
+                    itemCount: Array.isArray(data) ? data.length : 1,
+                    timestamp: Date.now(),
+                };
+
+            default:
+                return { task, data, executed: true };
+        }
+    }
+
+    /**
+     * Get pool status
+     */
+    getStatus() {
+        const workers = Array.from(this.workers.values());
+        return {
+            poolId: this.id,
+            status: this.status,
+            totalWorkers: workers.length,
+            idleWorkers: workers.filter(w => w.status === 'idle').length,
+            busyWorkers: workers.filter(w => w.status === 'busy').length,
+            virtualWorkers: workers.filter(w => w.isVirtual).length,
+            queuedTasks: this.taskQueue.length,
+            activeTasks: this.activeTasks.size,
+            stats: this.stats,
+        };
+    }
+
+    /**
+     * Shutdown the pool
+     */
+    async shutdown() {
+        this.status = 'shutting_down';
+
+        // Wait for active tasks
+        while (this.activeTasks.size > 0) {
+            await new Promise(r => setTimeout(r, 100));
+        }
+
+        // Clear workers
+        this.workers.clear();
+        this.status = 'shutdown';
+        this.emit('shutdown');
+    }
+}
+
+/**
+ * Task Orchestrator
+ *
+ * Orchestrates multi-agent workflows and complex task pipelines.
+ */
+export class TaskOrchestrator extends EventEmitter {
+    constructor(agentSpawner, workerPool, options = {}) {
+        super();
+        this.spawner = agentSpawner;
+        this.pool = workerPool;
+        this.workflows = new Map();
+        this.maxConcurrentWorkflows = options.maxConcurrentWorkflows || 5;
+    }
+
+    /**
+     * Create a workflow
+     */
+    createWorkflow(name, steps) {
+        const workflow = {
+            id: `wf-${randomBytes(6).toString('hex')}`,
+            name,
+            steps,
+            status: 'created',
+            currentStep: 0,
+            results: [],
+            startTime: null,
+            endTime: null,
+        };
+
+        this.workflows.set(workflow.id, workflow);
+        return workflow;
+    }
+
+    /**
+     * Execute a workflow
+     */
+    async executeWorkflow(workflowId, input = {}) {
+        const workflow = this.workflows.get(workflowId);
+        if (!workflow) throw new Error('Workflow not found');
+
+        workflow.status = 'running';
+        workflow.startTime = Date.now();
+        workflow.input = input;
+
+        this.emit('workflow-start', { workflowId, name: workflow.name });
+
+        try {
+            let context = { ...input };
+
+            for (let i = 0; i < workflow.steps.length; i++) {
+                workflow.currentStep = i;
+                const step = workflow.steps[i];
+
+                this.emit('step-start', {
+                    workflowId,
+                    step: i,
+                    type: step.type,
+                    name: step.name,
+                });
+
+                const result = await this.executeStep(step, context);
+                workflow.results.push(result);
+
+                // Pass result to next step
+                context = { ...context, [step.name || `step${i}`]: result };
+
+                this.emit('step-complete', {
+                    workflowId,
+                    step: i,
+                    result,
+                });
+            }
+
+            workflow.status = 'completed';
+            workflow.endTime = Date.now();
+
+            this.emit('workflow-complete', {
+                workflowId,
+                duration: workflow.endTime - workflow.startTime,
+                results: workflow.results,
+            });
+
+            return {
+                success: true,
+                results: workflow.results,
+                context,
+            };
+
+        } catch (error) {
+            workflow.status = 'failed';
+            workflow.endTime = Date.now();
+            workflow.error = error.message;
+
+            this.emit('workflow-failed', { workflowId, error: error.message });
+
+            return {
+                success: false,
+                error: error.message,
+                failedStep: workflow.currentStep,
+            };
+        }
+    }
+
+    /**
+     * Execute a single workflow step
+     */
+    async executeStep(step, context) {
+        switch (step.type) {
+            case 'agent':
+                return this.executeAgentStep(step, context);
+
+            case 'parallel':
+                return this.executeParallelStep(step, context);
+
+            case 'pool':
+                return this.executePoolStep(step, context);
+
+            case 'condition':
+                return this.executeConditionStep(step, context);
+
+            case 'transform':
+                return this.executeTransformStep(step, context);
+
+            default:
+                throw new Error(`Unknown step type: ${step.type}`);
+        }
+    }
+
+    /**
+     * Execute an agent step
+     */
+    async executeAgentStep(step, context) {
+        const task = typeof step.task === 'function'
+            ? step.task(context)
+            : step.task;
+
+        const agent = await this.spawner.spawn({
+            type: step.agentType || 'researcher',
+            task,
+            maxRuv: step.maxRuv,
+            priority: step.priority,
+        });
+
+        return new Promise((resolve, reject) => {
+            agent.on('complete', resolve);
+            agent.on('error', reject);
+
+            // Simulate completion for now
+            setTimeout(() => {
+                agent.complete({
+                    task,
+                    result: `Completed: ${task}`,
+                    context,
+                });
+            }, 1000);
+        });
+    }
+
+    /**
+     * Execute parallel agents
+     */
+    async executeParallelStep(step, context) {
+        const promises = step.agents.map(agentConfig =>
+            this.executeAgentStep(agentConfig, context)
+        );
+
+        return Promise.all(promises);
+    }
+
+    /**
+     * Execute worker pool step
+     */
+    async executePoolStep(step, context) {
+        const data = typeof step.data === 'function'
+            ? step.data(context)
+            : step.data || context.data;
+
+        return this.pool.execute({
+            task: step.task,
+            data,
+            strategy: step.strategy || 'parallel',
+        });
+    }
+
+    /**
+     * Execute conditional step
+     */
+    async executeConditionStep(step, context) {
+        const condition = typeof step.condition === 'function'
+            ? step.condition(context)
+            : step.condition;
+
+        if (condition) {
+            return this.executeStep(step.then, context);
+        } else if (step.else) {
+            return this.executeStep(step.else, context);
+        }
+        return null;
+    }
+
+    /**
+     * Execute transform step
+     */
+    async executeTransformStep(step, context) {
+        return step.transform(context);
+    }
+
+    /**
+     * Get workflow status
+     */
+    getWorkflowStatus(workflowId) {
+        const workflow = this.workflows.get(workflowId);
+        if (!workflow) return null;
+
+        return {
+            id: workflow.id,
+            name: workflow.name,
+            status: workflow.status,
+            currentStep: workflow.currentStep,
+            totalSteps: workflow.steps.length,
+            progress: (workflow.currentStep / workflow.steps.length) * 100,
+            startTime: workflow.startTime,
+            endTime: workflow.endTime,
+            duration: workflow.endTime && workflow.startTime
+                ? workflow.endTime - workflow.startTime
+                : null,
+        };
+    }
+
+    /**
+     * List all workflows
+     */
+    listWorkflows() {
+        return Array.from(this.workflows.values()).map(w => ({
+            id: w.id,
+            name: w.name,
+            status: w.status,
+            steps: w.steps.length,
+        }));
+    }
+}
+
+// Export default instances
+export default {
+    AGENT_TYPES,
+    TaskStatus,
+    DistributedAgent,
+    AgentSpawner,
+    WorkerPool,
+    TaskOrchestrator,
+};

package/package.json

@@ -1,8 +1,8 @@
 {
   "name": "@ruvector/edge-net",
-  "version": "0.1.2",
+  "version": "0.1.3",
   "type": "module",
-  "description": "Distributed compute intelligence network - contribute browser compute, earn credits. Features Time Crystal coordination, Neural DAG attention, and P2P swarm intelligence.",
+  "description": "Distributed compute intelligence network with AI agents and workers - contribute browser compute, spawn distributed AI agents, earn credits. Features Time Crystal coordination, Neural DAG attention, P2P swarm intelligence, and multi-agent workflows.",
   "main": "ruvector_edge_net.js",
   "module": "ruvector_edge_net.js",
   "types": "ruvector_edge_net.d.ts",
@@ -29,7 +29,13 @@
     "time-crystal",
     "dag-attention",
     "swarm-intelligence",
-    "neural-network"
+    "neural-network",
+    "ai-agents",
+    "distributed-agents",
+    "worker-pools",
+    "multi-agent",
+    "webrtc",
+    "task-orchestration"
   ],
   "author": "RuVector Team <team@ruvector.dev>",
   "license": "MIT",
@@ -54,6 +60,7 @@
     "network.js",
     "networks.js",
     "webrtc.js",
+    "agents.js",
     "README.md",
     "LICENSE"
   ],
@@ -64,6 +71,15 @@
     },
     "./wasm": {
       "import": "./ruvector_edge_net_bg.wasm"
+    },
+    "./agents": {
+      "import": "./agents.js"
+    },
+    "./network": {
+      "import": "./network.js"
+    },
+    "./webrtc": {
+      "import": "./webrtc.js"
     }
   },
   "sideEffects": [