agents-tab.md · 11.8 KB

Agents Tab: Agent Registry & AGIVITY Monitoring

Overview

The Agents Tab provides comprehensive agent management with real-time AGIVITY (AGI Activity) monitoring, cryptographic identity verification, and interactive agent cards for the mindX autonomous intelligence platform.

Status: ✅ DEPLOYED & OPERATIONAL Features: Agent cards, public key display, AGIVITY monitoring, RAGE memory integration Performance: Real-time updates with 15-second refresh intervals


🎯 Dashboard Sections

1. Agent Overview Statistics

Location: Top section with summary cards Metrics Displayed:
  • Total Agents: Complete agent count across all categories
  • Active Agents: Currently running and responsive agents
  • Registered Agents: Agents with cryptographic identities
  • Sovereign Agents: Agents with public keys for blockchain operations
  • 2. Agent Hierarchy Filter

    Location: Tab navigation below overview Categories:
  • All: Complete agent listing
  • System: Core system agents (Guardian, IDManager, Memory)
  • Orchestration: Coordination agents (Mastermind, Coordinator, CEO)
  • Intelligence: Cognitive agents (AGInt, BDI, Reasoning)
  • Specialized: Domain-specific agents (SimpleCoder, Blueprint, SEA)
  • User: User-created and custom agents
  • 3. Agent Cards Grid

    Location: Main content area Card Features:
  • Agent Identity: Name, ID, and cryptographic address
  • Status Indicator: Real-time health and activity status
  • Capability Tags: Skills and specializations
  • Public Key Badge: Visual indicator for cryptographic registration
  • Memory Badge: Count of stored memories for context awareness
  • Action Buttons: View details, interact, configure
  • 4. AGIVITY Monitor

    Location: Right sidebar panel Real-time Monitoring:
  • P-O-D-A Cognitive Loop: Perceive-Orient-Decide-Act visualization
  • Activity Stream: Live reasoning and decision logs
  • Loop Metrics: Cycle counts, average times, confidence scores
  • Q-Learning Updates: Reinforcement learning activity

  • 🤖 Agent Card Details

    Standard Agent Card

    ┌─────────────────────────────────────────┐
    │ 🤖 mindXagent                     🔑 ✓  │
    │ ID: mindxagent_main                     │
    │ Status: ● ACTIVE                        │
    ├─────────────────────────────────────────┤
    │ Type: orchestration                     │
    │ Capabilities: self-improvement,         │
    │   memory-feedback, campaign-management  │
    ├─────────────────────────────────────────┤
    │ Memory: 127 entries     Uptime: 99.9%   │
    │ [View Details] [Interact] [Configure]   │
    └─────────────────────────────────────────┘
    

    Card Components

    Identity Section

  • Agent Name: Display name for the agent
  • Agent ID: Unique identifier in the system
  • Public Key Indicator: Shows if agent has cryptographic identity
  • Status Light: Green (active), Yellow (degraded), Red (offline)
  • Capabilities Section

  • Agent Type: Category classification
  • Skills List: Specific capabilities and specializations
  • Tool Access: Authorized tools and resources
  • Metrics Section

  • Memory Count: Stored memories for context
  • Uptime: Availability percentage
  • Performance Score: Efficiency rating

  • 📊 AGIVITY Monitor

    P-O-D-A Cognitive Loop Visualization

    Perceive Phase

    🔍 PERCEIVE
    ├── Status: Active
    ├── Activity: Monitoring environment
    ├── Inputs: System events, user queries
    └── Duration: 0.3s average
    

    Orient Phase

    🧭 ORIENT
    ├── Status: Processing
    ├── Activity: Context building
    ├── Analysis: Pattern recognition
    └── Duration: 0.5s average
    

    Decide Phase

    🎯 DECIDE
    ├── Status: Thinking
    ├── Activity: Strategic analysis
    ├── Options: Evaluated alternatives
    └── Duration: 1.2s average
    

    Act Phase

    ⚡ ACT
    ├── Status: Executing
    ├── Activity: Task delegation
    ├── Actions: Dispatched commands
    └── Duration: 0.8s average
    

    Activity Stream

    Real-time log of agent cognitive processes:
    [14:32:15] mindXagent: Perceiving system state changes
    [14:32:16] mindXagent: Orienting on performance optimization
    [14:32:17] mindXagent: Deciding on memory consolidation strategy
    [14:32:18] mindXagent: Acting - Initiating RAGE semantic search
    [14:32:19] mindXagent: Completed cycle #1247 (3.2s, confidence: 94.7%)
    

    Loop Metrics

  • Total Cycles: Cumulative cognitive loop executions
  • Average Cycle Time: Mean duration per P-O-D-A cycle
  • Decision Confidence: Average certainty of decisions
  • Q-Learning Updates: Reinforcement learning iterations

  • 🔧 Technical Implementation

    Frontend Architecture

    Component Structure

    class AgentsTab extends TabComponent {
        constructor(config) {
            super({
                id: 'agents',
                label: 'Agents',
                refreshInterval: 15000, // 15-second updates
                autoRefresh: true
            });
        }
    }
    

    Data Integration

    // Agent registry data expression
    window.dataExpressions.registerExpression('agents_registry', {
        endpoints: [
            { url: '/agents', key: 'agents' },
            { url: '/registry/agents', key: 'registry' },
            { url: '/agents/keys', key: 'public_keys' }
        ],
        transform: (data) => this.transformAgentsData(data),
        onUpdate: (data) => this.updateAgentsRegistry(data)
    });

    // RAGE memory integration window.dataExpressions.registerExpression('rage_memory', { endpoints: [ { url: '/api/rage/memory/retrieve', method: 'POST', body: { query: 'agent activity', top_k: 20 } } ], transform: (data) => this.transformRageMemoryData(data), onUpdate: (data) => this.updateRageMemories(data) });

    Backend Endpoints

    Agent Registry

    GET /agents
    Response: {
        "agents": [
            {
                "agent_id": "mindxagent_main",
                "type": "orchestration",
                "status": "active",
                "capabilities": [...]
            }
        ]
    }
    

    Public Keys

    GET /agents/keys
    Response: {
        "mindxagent_main": "0x1234...abcd",
        "guardian_agent": "0xabcd...5678"
    }
    

    AGIVITY Stream

    GET /agi/activity/stream
    Response: {
        "activities": [
            {
                "timestamp": "2026-01-23T14:32:15Z",
                "agent_id": "mindxagent",
                "phase": "perceive",
                "activity": "Monitoring environment"
            }
        ]
    }
    

    🔐 Cryptographic Identity

    Public Key Display

    Each agent card shows cryptographic identity status:
  • 🔑 Verified: Agent has registered public key
  • ⚠️ Pending: Registration in progress
  • ❌ Unregistered: No cryptographic identity
  • Identity Verification Flow

    1. Agent Creation → IDManager.create_new_wallet()
    
  • Registry Registration → Guardian.validate_identity()
  • Public Key Assignment → blockchain_ready_identity
  • UI Display → Agent card public key badge
  • Wallet Integration

    // Verify agent cryptographic identity
    async verifyAgentIdentity(agentId) {
        const keyResponse = await this.apiRequest(/agents/keys/${agentId});
        return {
            hasKey: !!keyResponse.public_key,
            address: keyResponse.public_key,
            verified: keyResponse.verified,
            sovereign: keyResponse.blockchain_registered
        };
    }
    

    📚 RAGE Memory Integration

    Memory Context in Agent Cards

    Memory Badge Display

    updateAgentMemoryInsights(memories) {
        memories.forEach(memory => {
            const agentCard = document.querySelector([data-agent-id="${memory.agent_id}"]);
            if (agentCard) {
                // Add memory count badge
                const badge = agentCard.querySelector('.memory-badge');
                badge.textContent = ${memory.count} mem;
            }
        });
    }
    

    Semantic Memory Retrieval

    // Query RAGE for agent-specific memories
    async getAgentMemories(agentId, query) {
        return await this.apiRequest('/api/rage/memory/retrieve', {
            method: 'POST',
            body: {
                query: query,
                agent_id: agentId,
                top_k: 10,
                min_similarity: 0.5
            }
        });
    }
    

    Memory-Enhanced Agent Interactions

  • Context Loading: Retrieve relevant memories before agent interaction
  • History Display: Show memory timeline in agent detail modal
  • Similarity Search: Find related memories using semantic search
  • Memory Storage: Save interaction results for future context

  • 🎨 User Experience

    Visual Design

  • Cyberpunk Theme: Consistent with mindX platform aesthetic
  • Responsive Grid: Adapts to different screen sizes
  • Status Colors:
  • - 🟢 Green: Active/Healthy - 🟡 Yellow: Degraded/Warning - 🔴 Red: Offline/Error - 🔵 Blue: Processing/Thinking - 🟣 Purple: Learning/Improving

    Interaction Features

  • Hover Cards: Expanded information on hover
  • Click-through: Navigate to agent detail view
  • Quick Actions: Common actions directly on cards
  • Search/Filter: Find agents by name, type, or capability
  • Bulk Selection: Multi-agent operations
  • Accessibility

  • Keyboard Navigation: Tab through agent cards
  • Screen Reader: ARIA labels for all elements
  • High Contrast: Visible status indicators
  • Font Scaling: Responsive typography

  • 🔄 Real-Time Updates

    Refresh Intervals

  • Agent Status: 15-second updates
  • AGIVITY Stream: 1-second updates (live)
  • Memory Context: 30-second updates
  • Registry Data: 60-second updates
  • WebSocket Integration (Optional)

    // Real-time AGIVITY stream via WebSocket
    const socket = new WebSocket('ws://localhost:8000/ws/agivity');
    socket.onmessage = (event) => {
        const activity = JSON.parse(event.data);
        this.updateActivityStream(activity);
    };
    

    📊 Agent Statistics

    Overview Metrics

  • Total Agents Found: 30+ agents in codebase
  • Registered Agents: 9 with cryptographic identity
  • Active Agents: Real-time count of running agents
  • Sovereign Agents: Blockchain-ready agents
  • Performance Metrics

  • Average Response Time: Agent task completion latency
  • Success Rate: Task completion percentage
  • Memory Usage: Per-agent memory consumption
  • Uptime: Individual agent availability

  • 🐛 Troubleshooting

    Common Issues

    Agents Not Displaying

    # Check backend is running
    curl http://localhost:8000/agents

    Verify agent registry

    curl http://localhost:8000/registry/agents

    Check for JavaScript errors in browser console

    AGIVITY Not Updating

    # Check AGInt service status
    curl http://localhost:8000/agi/activity/stream

    Verify WebSocket connection (if used)

    Check browser console for connection errors

    Memory Badges Missing

    # Test RAGE memory endpoint
    curl -X POST http://localhost:8000/api/rage/memory/retrieve \
      -H "Content-Type: application/json" \
      -d '{"query": "agent activity", "top_k": 10}'
    

    📚 Related Documentation

  • RAGE System: Retrieval augmented generation
  • Identity Management: Cryptographic identity framework
  • Agent Registry: Complete agent catalog
  • Platform Tab: Enterprise dashboard
  • pgvectorscale Integration: Semantic memory

  • The Agents Tab provides comprehensive visibility into the mindX autonomous agent ecosystem, enabling effective monitoring, management, and interaction with all system agents.


    Referenced in this document
    AGENTSIDENTITYpgvectorscale_memory_integrationplatform-tab

    All DocumentsDocument IndexThe Book of mindXImprovement JournalAPI Reference