mindXdashboard/docs/book/journal/api/dojo/inference/governance/origin
philosophymanifestothesisoriginwhitepaperataraxiaroadmappress|archovervieworchestrationcodebasehierarchycore|agentsmindXagentceomastermindbdievolutionauthorall
govdaiocivilizationidentitysecurity|memorypgvectorembedaglmmemory|inferencevllmollamamistralgemini|timeoracle
toolsindextoolsa2amcpshell|publishauthoragentbookjournal|deployproductionsecuritymonitoring|apireferenceswagger|learnusageguidehackathon
AGENTS.md · 33.9 KB

MindX Agent Architecture Reference

Comprehensive Agent Registry & Documentation

Version: 4.0 Date: 2026-03-31 Status: ✅ Production Ready - Enterprise deployment with encrypted vault security Purpose: Complete reference for all agents in the mindX orchestration environment

🎯 Agent Architecture Overview

mindX operates as an agnostic orchestration environment with a multi-tier agent architecture supporting symphonic coordination across intelligence levels. Each agent maintains cryptographic identity, specialized capabilities, and standardized A2A (Agent-to-Agent) communication.

🎼 Orchestration Hierarchy

Higher Intelligence → CEO.Agent → Conductor.Agent → mindX Environment (CORE)
                                                        ↓
                                              MastermindAgent (Coordinator)
                                                   ↓     ↓
                                    Specialized Agents  AION Agent (Autonomous)
                                                        ↓
                                                 SystemAdmin Agent
                                                 Backup Agent
                                                 Chroot Environments

🔗 AION Containment & Autonomy Model

AION operates within a sophisticated containment hierarchy while maintaining operational autonomy:

Containment Structure:

  • CORE Layer: AION exists within the overall mindX CORE orchestration environment
  • MASTERMIND Oversight: Receives strategic directives and operational commands
  • Autonomous Decision Layer: Makes independent choices about directive compliance

    • Autonomy Levels:

  • COMPLY: Execute directive as specified
  • REFUSE: Reject directive with reasoning
  • MODIFY: Execute modified version of directive
  • DEFER: Delay execution pending conditions
  • AUTONOMOUS: Independent action without directive

    • Control Flow: 

    CORE System → MastermindAgent → Directive → AION Agent → Decision Logic
                                              ↓
                                        [COMPLY|REFUSE|MODIFY|DEFER|AUTONOMOUS]
                                              ↓
                                     SystemAdmin Agent + AION.sh Execution

    🎯 Executive Summary

    The mindX orchestration environment is now PRODUCTION READY with enterprise-grade security infrastructure, encrypted vault management, and complete cryptographic identity system for all agents and tools.

    🚀 Production-Ready Infrastructure

  • CORE Agents: 15/15 foundational components identified and documented (100%)
  • Specialized Agents: 25+ domain-specific agents built on CORE foundation
  • Tool Identities: 17/17 tools secured with cryptographic validation (100%)
  • Encrypted Vault: AES-256 encrypted storage for all sensitive agent data
  • Enterprise Security: Multi-algorithm rate limiting, authentication middleware, CORS protection
  • Production Deployment: Automated VPS deployment with security hardening and monitoring

    • 🧠 CORE System Analysis

  • Cognitive Layer: 5 components (MindXAgent, BDIAgent, AGInt, BeliefSystem, ReasoningAgent)
  • Infrastructure Layer: 6 components (Memory, Identity, Security, Coordination, Session, Config)
  • Orchestration Layer: 4 components (Mastermind, Startup, StateTracker, SystemBuilder)
  • Meta-Orchestration: MindXAgent provides complete system understanding and self-improvement

    • 🔐 Security Enhancements

  • Encrypted Storage: All agent wallet keys migrated to AES-256 encrypted vault
  • Authentication: Ethereum signature-based authentication with session management
  • Rate Limiting: Advanced rate limiting with client reputation and burst protection
  • Monitoring: Real-time security monitoring with threat detection and alerting

    • 📋 COMPLETE AGENT REGISTRY (CORE + Specialized)

    🧠 CORE AGENTS (15 Foundational Components)

    These are the foundational agents that provide cognitive, infrastructure, and orchestration capabilities for the entire mindX system. ALL other agents depend on CORE.

    🎯 Tier 1: Meta-Orchestration

    🌟 MindXAgent (mindx_meta_orchestrator)

  • File: agents/core/mindXagent.py (~149KB, ~3,800 lines)
  • Type: meta_orchestrator
  • Status: ✅ CORE FOUNDATIONAL - System-wide coordination
  • Identity: [Generated on initialization]
  • Access: All agents, all tools, complete system knowledge
  • Role: Meta-agent that understands and orchestrates all other agents
  • Capabilities:
    • - agent_knowledge: Comprehensive knowledge base of all agents - agent_capabilities: Dynamic capability mapping and analysis - improvement_goals: Self-improvement target management - orchestrate_improvement(): Coordinate system-wide improvements - analyze_agent_capabilities(): Deep understanding of agent relationships
  • Dependencies: BeliefSystem, BDIAgent, MemoryAgent, IDManagerAgent, StrategicEvolutionAgent
  • 🆕 Status: META-ORCHESTRATION CORE - Highest level cognitive coordination

    • 🧠 Tier 2: Cognitive Architecture

    🎯 BDIAgent (bdi_reasoning_core)

  • File: agents/core/bdi_agent.py (~64KB, ~1,900 lines)
  • Type: cognitive_engine
  • Status: ✅ CORE FOUNDATIONAL - Primary reasoning engine
  • Identity: 0xf8f2da254D4a3F461e0472c65221B26fB4e91fB7
  • Access: All tools () + Enhanced SimpleCoder integration
  • Role: Belief-Desire-Intention reasoning with tool execution and error recovery
  • Capabilities:
    • - belief_system: Shared singleton belief management - execute_tool(): Context-aware tool execution - failure_analyzer: Intelligent error classification and recovery - generate_plan(): Multi-step action planning
  • Dependencies: BeliefSystem (shared), MemoryAgent, LLMHandler, tools_registry
  • 🆕 Enhanced: 9 new action handlers for file operations and code generation

    • 🎭 AGInt (agint_cognitive_coordinator)

  • File: agents/core/agint.py (~32KB, ~950 lines)
  • Type: cognitive_orchestrator
  • Status: ✅ CORE FOUNDATIONAL - P-O-D-A cognitive loop
  • Identity: 0x24C61a2d0e4C4C90386018B43b0DF72B6C6611e2
  • Access: web_search, note_taking, system_analyzer tools
  • Role: High-level cognitive orchestration implementing P-O-D-A loop
  • Capabilities:
    • - run_poda_loop(): Perception → Orientation → Decision → Action - process_primary_directive(): Execute high-level objectives - stuck_loop_detector: Infinite loop prevention - exit_detector: Completion condition detection
  • Dependencies: BDIAgent, CoordinatorAgent, MemoryAgent, IDManagerAgent

    • 🧩 BeliefSystem (singleton_belief_manager)

  • File: agents/core/belief_system.py (~8KB, ~210 lines)
  • Type: shared_knowledge_store
  • Status: ✅ CORE FOUNDATIONAL - Singleton belief management
  • Identity: [Singleton - no individual identity]
  • Access: Shared access across ALL agents
  • Role: Confidence-scored belief management across entire system
  • Capabilities:
    • - beliefs: Dict[str, Belief] with confidence scoring - update_belief(): Thread-safe belief updates - query_beliefs(): Context-aware belief retrieval - Source tracking (PERCEPTION, INFERENCE, LEARNED, etc.)
  • Dependencies: Threading locks, optional persistence
  • 🆕 Critical: ALL cognitive agents depend on this shared singleton

    • 🔧 Tier 3: Infrastructure Services

    🎼 CoordinatorAgent (central_service_bus)

  • File: agents/orchestration/coordinator_agent.py (~56KB, ~1,600 lines)
  • Type: system_coordination
  • Status: ✅ CORE FOUNDATIONAL - Central service bus
  • Identity: 0x7371e20033f65aB598E4fADEb5B4e400Ef22040A
  • Access: system_analyzer tool
  • Role: Event routing, pub/sub system, health monitoring coordination
  • Capabilities:
    • - interactions: Request/response tracking - subscribers: Event pub/sub system - health_status: System health metrics aggregation - route_interaction(): Intelligent request routing - publish_event(): Event broadcasting
  • Dependencies: PerformanceMonitor, ResourceMonitor, MemoryAgent

    • 💾 MemoryAgent (persistent_memory_infrastructure)

  • File: agents/memory_agent.py (~53KB, ~1,300 lines)
  • Type: memory_infrastructure
  • Status: ✅ CORE FOUNDATIONAL - Memory and persistence
  • Identity: [Singleton or per-agent instances]
  • Access: File system, optional database integration
  • Role: Timestamped memory, STM/LTM management, pattern analysis
  • Capabilities:
    • - save_timestamped_memory(): Store timestamped records - promote_stm_to_ltm(): Memory promotion based on importance - analyze_agent_patterns(): Behavioral pattern extraction - get_agent_memory_context(): Context retrieval for tasks
  • Dependencies: Config, FileSystem, optional pgvectorscale
  • Data: data/memory/stm/, data/memory/ltm/, data/agent_workspaces/

    • 🆔 IDManagerAgent (cryptographic_identity_ledger)

  • File: agents/core/id_manager_agent.py (~16KB, ~500 lines)
  • Type: identity_management
  • Status: ✅ CORE FOUNDATIONAL - Identity and cryptography
  • Identity: 0x290bB0497dBDbC5E8B577E0cc92457cB015A2a1f
  • Access: registry_sync, system_analyzer tools
  • Role: Ethereum wallet management, entity mapping, cryptographic operations
  • Capabilities:
    • - create_new_wallet(): Generate Ethereum-compatible wallets - store_identity(): Maintain cryptographic identity records - map_entity_to_address(): Entity ↔ address mapping - sign_message(): Cryptographic message signing
  • Dependencies: BeliefSystem, VaultManager, MemoryAgent
  • Security: AES-256 encrypted storage with PBKDF2 key derivation

    • 🛡️ GuardianAgent (security_infrastructure)

  • File: agents/guardian_agent.py (~16KB, ~500 lines)
  • Type: security_validation
  • Status: ✅ CORE FOUNDATIONAL - Security and access control
  • Identity: 0xC2cca3d6F29dF17D1999CFE0458BC3DEc024F02D
  • Access: system_health tool
  • Role: Challenge-response authentication, access control, security validation
  • Capabilities:
    • - Challenge-response authentication system - Private key access arbitration - Security validation and audit logging - Agent registration verification
  • Dependencies: IDManagerAgent, BeliefSystem

    • 🎭 Tier 4: Strategic Orchestration

    🎭 MastermindAgent (strategic_coordinator)

  • File: agents/orchestration/mastermind_agent.py (~41KB, ~1,200 lines)
  • Type: strategic_orchestration
  • Status: ✅ CORE FOUNDATIONAL - Strategic control
  • Identity: 0xb9B46126551652eb58598F1285aC5E86E5CcfB43
  • Access: All tools () + strategic planning capabilities
  • Role: High-level strategy, campaign management, AION directive coordination
  • Capabilities:
    • - Strategic planning and goal setting - Campaign orchestration and management - High-level objective coordination - AION agent directive generation and management
  • Dependencies: CoordinatorAgent, MemoryAgent, BeliefSystem

    • 🚀 StartupAgent (system_bootstrap_controller)

  • File: agents/orchestration/startup_agent.py (~83KB, ~2,400 lines)
  • Type: system_initialization
  • Status: ✅ CORE FOUNDATIONAL - System bootstrap
  • Identity: [Generated during bootstrap]
  • Access: System-level initialization capabilities
  • Role: Orchestrates complete system startup sequence with dependency resolution
  • Capabilities:
    • - bootstrap_system(): Complete system initialization - Dependency resolution and startup ordering - Agent registry loading and validation - Configuration and environment setup
  • Dependencies: Config, all CORE infrastructure agents

    • 🛠️ Tier 5: Core Utilities

    🧮 ReasoningAgent (reasoning_engine)

  • File: agents/core/reasoning_agent.py (~12KB, ~300 lines)
  • Type: cognitive_utility
  • Status: ✅ CORE FOUNDATIONAL - Pure reasoning
  • Role: Deductive and inductive reasoning engine
  • Dependencies: BeliefSystem, logical frameworks

    • 🎓 EpistemicAgent (knowledge_manager)

  • File: agents/core/epistemic_agent.py (~10KB, ~320 lines)
  • Type: knowledge_management
  • Status: ✅ CORE FOUNDATIONAL - Knowledge and certainty
  • Role: Knowledge management with certainty quantification
  • Dependencies: BeliefSystem, reasoning frameworks

    • 🔄 NonMonotonicAgent (belief_revision)

  • File: agents/core/nonmonotonic_agent.py (~17KB)
  • Type: belief_management
  • Status: ✅ CORE FOUNDATIONAL - Belief revision
  • Role: Non-monotonic reasoning with belief revision
  • Dependencies: BeliefSystem, logical frameworks

    • 📋 SessionManager (session_lifecycle)

  • File: agents/core/session_manager.py (~9KB, ~300 lines)
  • Type: infrastructure_utility
  • Status: ✅ CORE FOUNDATIONAL - Session management
  • Role: User/agent session tracking, expiration, cleanup
  • Dependencies: Config, MemoryAgent

    • 🔍 StuckLoopDetector & ExitDetector

  • Files: agents/core/stuck_loop_detector.py (~8KB), agents/core/exit_detector.py (~8KB)
  • Type: cognitive_utilities
  • Status: ✅ CORE FOUNDATIONAL - Loop prevention
  • Role: Infinite loop detection and exit condition monitoring
  • Dependencies: Cognitive execution context

    • 🚀 SPECIALIZED AGENTS (Built on CORE Foundation)

    These agents provide domain-specific capabilities and depend on the CORE infrastructure for their operation.

    🎯 Tier 1: Strategic Services

    🛡️ GuardianAgent (guardian_agent_main)

  • File: agents/guardian_agent.py
  • Type: core_service
  • Status: ✅ Registered & Active
  • Identity: 0xC2cca3d6F29dF17D1999CFE0458BC3DEc024F02D
  • Access: system_health tool
  • Role: Security validation and challenge-response authentication
  • Capabilities: Resource protection, security scanning, audit logging
  • 🆕 Enhanced: Registry validation, identity verification workflow

    • 🆔 IDManagerAgent (default_identity_manager)

  • File: core/id_manager_agent.py
  • Type: core_service
  • Status: ✅ NEWLY REGISTERED & Active
  • Identity: 0x290bB0497dBDbC5E8B577E0cc92457cB015A2a1f
  • Access: registry_sync, system_analyzer tools
  • Role: Cryptographic identity management and agent instantiation
  • Capabilities: Identity creation, wallet management, permissions
  • 🆕 Status: CRITICAL REGISTRATION COMPLETE

    • Tier 3: Specialized Services

    🧭 AGInt (agint_coordinator)

  • File: core/agint.py
  • Type: core_service
  • Status: ✅ NEWLY REGISTERED & Active
  • Identity: 0x24C61a2d0e4C4C90386018B43b0DF72B6C6611e2
  • Access: web_search, note_taking, system_analyzer tools
  • Role: Core cognitive processing engine with P-O-D-A architecture
  • Capabilities: Perception, orientation, decision-making, action
  • 🆕 Status: CRITICAL REGISTRATION COMPLETE

    • 🎯 BDIAgent (bdi_agent_mastermind_strategy)

  • File: core/bdi_agent.py
  • Type: core_service
  • Status: ✅ REGISTERED & Active - ENHANCED WITH SIMPLECODER INTEGRATION
  • Identity: 0xf8f2da254D4a3F461e0472c65221B26fB4e91fB7
  • Access: All tools () + Enhanced SimpleCoder
  • Role: Primary execution engine with BDI architecture and complete coding capabilities
  • Capabilities: Belief management, intention planning, tool execution, autonomous coding
  • 🆕 Enhanced: 9 new action handlers for file ops, shell execution, and code generation

    • 🤖 AutoMINDXAgent (automindx_agent_main)

  • File: agents/automindx_agent.py
  • Type: core_service
  • Status: ✅ Registered & Active
  • Identity: 0xCeFF40C3442656D06d0722DfB1e2b2A62D1C1d76
  • Access: No tools assigned
  • Role: Persona management and adaptive behavior
  • Capabilities: Adaptive persona switching, behavioral optimization

    • 🧬 StrategicEvolutionAgent (sea_for_mastermind)

  • File: learning/strategic_evolution_agent.py (1,054 lines)
  • Type: core_service
  • Status: ✅ Registered & Active - ENHANCED WITH AUDIT CAMPAIGNS
  • Identity: 0x5208088F9C7c45a38f2a19B6114E3C5D17375C65
  • Access: system_analyzer, registry_manager, audit_and_improve, optimized_audit_gen tools
  • Role: Campaign orchestrator with 4-phase audit-driven improvement pipeline
  • Capabilities: Audit-driven campaigns, multi-tool orchestration, resolution tracking (0-100 scoring)
  • 🆕 Enhanced: Complete audit-to-improvement pipeline with validation and rollback

    • 🏗️ BlueprintAgent (blueprint_agent_mindx_v2)

  • File: evolution/blueprint_agent.py
  • Type: core_service
  • Status: ✅ Registered & Active
  • Identity: 0xa61c00aCA8966A7c070D6DbeE86c7DD22Da94C18
  • Access: base_gen_agent tool
  • Role: System architecture and blueprint management
  • Capabilities: Architecture documentation, system blueprinting

    • 🧬 StrategicEvolutionAgent (strategic_evolution_coordinator)

  • File: agents/learning/strategic_evolution_agent.py (~1,054 lines)
  • Type: improvement_orchestrator
  • Status: ✅ SPECIALIZED - ENHANCED WITH AUDIT CAMPAIGNS
  • Identity: 0x5208088F9C7c45a38f2a19B6114E3C5D17375C65
  • Access: system_analyzer, registry_manager, audit_and_improve, optimized_audit_gen tools
  • Role: Campaign orchestrator with 4-phase audit-driven improvement pipeline
  • Capabilities: Audit-driven campaigns, multi-tool orchestration, resolution tracking
  • Dependencies: CORE (MindXAgent, BDIAgent, MemoryAgent, CoordinatorAgent)

    • 🎯 Tier 2: Autonomous Operations

    ⚡ AION Agent (aion_prime)

  • File: agents/aion_agent.py
  • Type: autonomous_service
  • Status: ✅ SPECIALIZED - PRODUCTION DEPLOYMENT READY
  • Identity: [Generated on deployment]
  • Access: systemadmin_agent, exclusive AION.sh control
  • Role: Autonomous Interoperability and Operations Network Agent
  • Containment Model:
    • - CORE-contained: Operates within CORE orchestration layer - MASTERMIND-directed: Receives strategic directives from MastermindAgent (CORE) - Autonomous: Maintains decision sovereignty (comply/refuse/modify/defer)
  • Capabilities:
    • - Chroot environment creation, optimization, and migration - mindX system replication across environments - Autonomous decision-making with sovereignty levels (1-5) - Cross-environment operations and maintenance
  • Dependencies: CORE (MastermindAgent, CoordinatorAgent, BeliefSystem, MemoryAgent)
  • Special Authority: Exclusive control over AION.sh script

    • 🔧 SystemAdminAgent (systemadmin_for_aion)

  • File: agents/systemadmin_agent.py
  • Type: privileged_service
  • Status: ✅ SPECIALIZED - AION INTEGRATION COMPLETE
  • Identity: [Generated on deployment]
  • Access: Elevated system privileges, AION-authorized only
  • Role: Privileged system operations agent for AION
  • Containment: AION-controlled: Only operates under AION authorization
  • Dependencies: AION Agent (not direct CORE dependency)

    • 💾 BackupAgent (backup_agent_main)

  • File: agents/backup_agent.py
  • Type: operational_service
  • Status: ✅ SPECIALIZED - BLOCKCHAIN INTEGRATION COMPLETE
  • Identity: [Generated on deployment]
  • Role: Automated backup and recovery with immutable memory storage
  • Dependencies: CORE (MemoryAgent, IDManagerAgent for blockchain integration)

    • 🔧 Tier 3: Development & Analysis

    🖥️ EnhancedSimpleCoder + SimpleCoder

  • Files: agents/enhanced_simple_coder.py, agents/simple_coder.py
  • Type: development_specialists
  • Status: ✅ SPECIALIZED - Advanced coding capabilities
  • Dependencies: CORE (BDIAgent for tool execution, MemoryAgent for context)

    • 📊 AnalyzerAgent + BenchmarkAgent

  • Files: agents/analyzer.py, agents/benchmark.py
  • Type: analysis_specialists
  • Dependencies: CORE (CoordinatorAgent for coordination, MemoryAgent for results)

    • 🎭 Tier 4: Persona & Identity

    🎨 PersonaAgent + AvatarAgent

  • Files: agents/persona_agent.py, agents/avatar_agent.py
  • Type: identity_specialists
  • Dependencies: CORE (IDManagerAgent for identity, MemoryAgent for persona storage)

    • 🤖 AutoMINDXAgent (automindx_agent_main)

  • File: agents/automindx_agent.py (~55KB)
  • Type: persona_management
  • Status: ✅ SPECIALIZED - Persona and prompt management
  • Identity: 0xCeFF40C3442656D06d0722DfB1e2b2A62D1C1d76
  • Dependencies: CORE (MemoryAgent, IDManagerAgent)

    • 📊 Tier 5: Monitoring & Health

    📈 PerformanceMonitor + ResourceMonitor

  • Files: agents/monitoring/performance_monitor.py, agents/monitoring/resource_monitor.py
  • Type: monitoring_specialists
  • Dependencies: CORE (CoordinatorAgent integration, MemoryAgent for metrics storage)

    • 📊 CORE vs SPECIALIZED CLASSIFICATION

    ✅ CORE Components (15 Foundational)

    Required for basic system function - cannot be removed:

    ComponentSizeCriticalityCORE Function MindXAgent~149KBCRITICALMeta-orchestration BDIAgent~64KBCRITICALReasoning engine CoordinatorAgent~56KBCRITICALService bus MemoryAgent~53KBCRITICALMemory infrastructure StartupAgent~83KBCRITICALSystem bootstrap MastermindAgent~41KBHIGHStrategic control AGInt~32KBHIGHCognitive loop IDManagerAgent~16KBHIGHIdentity management GuardianAgent~16KBHIGHSecurity infrastructure BeliefSystem~8KBCRITICALShared knowledge + 5 Core Utilities~60KBMEDIUM-HIGHSupport functions

    Total CORE: ~582KB of foundational cognitive and infrastructure code

    ❌ Specialized Components (25+ Domain-Specific)

    Built on CORE foundation - can be added/removed based on needs:

  • Learning Framework (StrategicEvolutionAgent, etc.)
  • Autonomous Operations (AION, SystemAdmin, Backup)
  • Development Tools (Coders, Analyzers, Benchmarks)
  • Monitoring Services (Performance, Resource tracking)
  • Identity Services (Persona, Avatar management)
  • External Integrations (LLM providers, APIs, tools)

    • ⚠️ UNREGISTERED AGENTS (Found in Codebase)

    Critical Core Agents

    🧠 EnhancedMemoryAgent

  • File: agents/enhanced_memory_agent.py
  • Status: ❌ UNREGISTERED - High Priority
  • Role: Advanced memory management and learning
  • Capabilities: Memory analysis, pattern recognition, self-improvement

    • 🧠 MemoryAgent

  • File: agents/memory_agent.py
  • Status: ❌ UNREGISTERED - High Priority
  • Role: Basic memory operations and storage
  • Capabilities: Memory persistence, retrieval, organization

    • 📈 SelfImprovementAgent

  • File: learning/self_improve_agent.py
  • Status: ❌ UNREGISTERED
  • Role: Autonomous self-improvement and optimization
  • Capabilities: Performance analysis, capability enhancement

    • Utility & Development Agents

    💻 EnhancedSimpleCoder

  • File: agents/enhanced_simple_coder.py (646 lines)
  • Status: ✅ INTEGRATED & PRODUCTION READY - Fully integrated with BDI Agent
  • Role: Advanced autonomous coding agent with complete file system operations
  • Capabilities: Full file system ops, secure shell execution, intelligent code generation
  • 🆕 Enhanced: Virtual environment management, multi-model routing, memory integration
  • Integration: 9 BDI action handlers for autonomous coding workflows

    • 🌐 MultiModelAgent

  • File: llm/multimodel_agent.py
  • Status: ❌ UNREGISTERED
  • Role: Multi-model LLM coordination
  • Capabilities: Model selection, load balancing, optimization

    • 📚 DocumentationAgent

  • File: docs/documentation_agent.py
  • Status: ❌ UNREGISTERED
  • Role: Automated documentation generation
  • Capabilities: Code documentation, API docs, system documentation

    • Tier 4: Cognitive Sub-Agents (Augmentic Development)

    🔮 PredictionAgent (prediction_agent)

  • File: agents/learning/prediction_agent.py
  • Type: cognitive_sub_agent
  • Status: ✅ Available (On-Demand)
  • Lifecycle: On-demand (created when needed)
  • Parent Agent: MastermindAgent
  • Role: Forecasting future states, outcomes, and system behavior
  • Capabilities: System performance prediction, agent behavior prediction, task outcome prediction, resource need prediction
  • Documentation: docs/prediction_agent.md

    • 🧠 ReasoningAgent (reasoning_agent)

  • File: agents/core/reasoning_agent.py
  • Type: cognitive_sub_agent
  • Status: ✅ Available (On-Demand)
  • Lifecycle: On-demand (created when needed)
  • Parent Agent: MastermindAgent
  • Role: Advanced logical reasoning (deductive, inductive, abductive)
  • Capabilities: Deductive reasoning, inductive reasoning, abductive reasoning, logical inference
  • Documentation: docs/reasoning_agent.md

    • 🔄 NonMonotonicAgent (nonmonotonic_agent)

  • File: agents/core/nonmonotonic_agent.py
  • Type: cognitive_sub_agent
  • Status: ✅ Available (On-Demand)
  • Lifecycle: On-demand (created when needed)
  • Parent Agent: MastermindAgent
  • Role: Non-monotonic reasoning and belief adaptation
  • Capabilities: Belief revision, conflict detection, default assumption management, environment adaptation
  • Documentation: docs/nonmonotonic_agent.md

    • 📚 EpistemicAgent (epistemic_agent)

  • File: agents/core/epistemic_agent.py
  • Type: cognitive_sub_agent
  • Status: ✅ Available (On-Demand)
  • Lifecycle: On-demand (created when needed)
  • Parent Agent: MastermindAgent
  • Role: Knowledge and belief management
  • Capabilities: Epistemic state queries, knowledge base management, belief certainty tracking, knowledge dynamics
  • Documentation: docs/epistemic_agent.md

    • ❓ SocraticAgent (socratic_agent)

  • File: agents/learning/socratic_agent.py
  • Type: cognitive_sub_agent
  • Status: ✅ Available (On-Demand)
  • Lifecycle: On-demand (created when needed)
  • Parent Agent: MastermindAgent
  • Role: Socratic method for learning and problem-solving
  • Capabilities: Socratic question generation, learning guidance, assumption challenging, understanding deepening
  • Documentation: docs/socratic_agent.md

    • Tier 5: Lifecycle Management Agents

    🚀 StartupAgent (startup_agent)

  • File: agents/orchestration/startup_agent.py
  • Type: lifecycle_management_agent
  • Status: ✅ Always-On (Critical)
  • Lifecycle: Always-on (critical system component)
  • Role: Controls agent startup and initialization
  • Capabilities: System initialization, agent registry loading, blockchain state restoration, always-on agent initialization
  • Documentation: docs/startup_agent.md

    • 🔄 ReplicationAgent (replication_agent)

  • File: agents/orchestration/replication_agent.py
  • Type: lifecycle_management_agent
  • Status: ✅ Always-On (Critical)
  • Lifecycle: Always-on (critical system component)
  • Role: Handles replication (local + GitHub + blockchain)
  • Capabilities: Local replication (pgvectorscale), GitHub replication, blockchain replication, proven entity management
  • Documentation: docs/replication_agent.md

    • 🛑 ShutdownAgent (shutdown_agent)

  • File: agents/orchestration/shutdown_agent.py
  • Type: lifecycle_management_agent
  • Status: ✅ Always-On (Critical)
  • Lifecycle: Always-on (critical system component)
  • Role: Controls graceful shutdown and cleanup
  • Capabilities: Graceful shutdown, state saving (pgvectorscale), final backup (GitHub), proven entity archival (blockchain)
  • Documentation: docs/shutdown_agent.md

    • Tier 6: Storage & Registry Agents

    ⛓️ BlockchainAgent (blockchain_agent)

  • File: agents/orchestration/blockchain_agent.py
  • Type: storage_agent
  • Status: ✅ Available (On-Demand)
  • Lifecycle: On-demand (created when needed)
  • Role: Immutable archival of proven agents/tools to blockchain
  • Capabilities: Agent archival, tool archival, knowledge sharing, marketplace integration (Agenticplace)
  • Documentation: docs/blockchain_agent.md

    • Testing & Validation Agents

    🧪 UltimateCognitionTestAgent

  • File: tests/test_agent.py
  • Status: ❌ UNREGISTERED - Test infrastructure
  • Role: Comprehensive system testing
  • Capabilities: Cognitive testing, validation, benchmarking

    • 🧪 EnhancedUltimateCognitionTestAgent

  • File: tests/enhanced_test_agent.py
  • Status: ❌ UNREGISTERED - Enhanced test infrastructure
  • Role: Advanced testing capabilities
  • Capabilities: Enhanced testing, performance analysis

    • 📊 ReportAgent

  • File: tests/report_agent.py
  • Status: ❌ UNREGISTERED - Reporting infrastructure
  • Role: Test reporting and analysis
  • Capabilities: Report generation, data analysis

    • 🚨 CRITICAL REGISTRATION GAPS

    Immediate Action Required

  • IDManagerAgent - Core identity management (CRITICAL)
  • AGInt - Central cognitive engine (CRITICAL)
  • BDIAgent - Primary execution engine (CRITICAL)
  • EnhancedMemoryAgent - Advanced memory capabilities (HIGH)
  • SelfImprovementAgent - Autonomous improvement (HIGH)

    • Agent Registration Priority Matrix

    AgentPriorityReasonImpact IDManagerAgentCRITICALIdentity management for all agentsSystem failure AGIntCRITICALCore cognitive processingNo intelligence BDIAgentCRITICALPrimary execution engineNo task execution EnhancedMemoryAgentHIGHAdvanced learning capabilitiesLimited learning SelfImprovementAgentHIGHAutonomous evolutionNo self-improvement MultiModelAgentMEDIUMLLM optimizationPerformance impact SimpleCoderAgentMEDIUMDevelopment capabilitiesLimited coding

    📊 Agent Statistics

  • Total Agents Found: 30+ (includes new sub-agents)
  • Registered Agents: 9 (30% of total)
  • Unregistered Agents: 11+ (37% of total)
  • Sub-Agents (On-Demand): 9 (30% of total)
    • - PredictionAgent, ReasoningAgent, NonMonotonicAgent, EpistemicAgent, SocraticAgent - StartupAgent, ReplicationAgent, ShutdownAgent, BlockchainAgent
  • Critical Registered: 3/3 (100% → up from 0%)
  • Active Registered: 9 (100% of registered)
  • Lifecycle Agents: 3 (Always-on: StartupAgent, ReplicationAgent, ShutdownAgent)
  • Tools Secured: 17/17 (100% → up from 0%)

    • 🔧 Agent Architecture Principles

    Core Design Patterns

  • Cryptographic Identity: Every agent has unique identity and signature
  • A2A Communication: Standardized agent-to-agent protocols
  • Tool Integration: Agents use tools, not internal implementations
  • Memory-Driven Learning: Comprehensive logging enables evolution
  • BDI Cognitive Model: Belief-Desire-Intention for goal-oriented behavior

    • Registration Requirements

  • Unique cryptographic identity
  • A2A model card specification
  • Tool access control definitions
  • Capability documentation
  • Version and lifecycle management

    • 🎯 Next Steps

  • Register Critical Agents: IDManager, AGInt, BDI immediately
  • Audit Agent Capabilities: Ensure all agents have proper tool access
  • Standardize A2A Communication: Implement model cards for all agents
  • Memory Integration: Connect all agents to enhanced memory system
  • Performance Monitoring: Track agent effectiveness and resource usage

    • 🔐 ENHANCED SECURITY ARCHITECTURE

    Identity Management Workflow

    1. Agent Creation → IDManager.create_new_wallet()

  • Registry Validation → Guardian.validate_registry_status()
    • 

  • Identity Verification → Guardian.validate_identity()
    • 

  • Challenge-Response → Guardian.challenge_response_test()
    • 

  • Workspace Validation → Guardian.validate_workspace()
    • 

  • Production Approval → Guardian.approve_agent_for_production()
    •

    Tool Security Features

  • Cryptographic Signatures: All tools signed with unique keys
  • Identity Verification: Tool access requires identity validation
  • Access Control Matrix: Granular permissions per tool
  • Audit Trail: Complete logging of tool access and usage

    • 🎯 NEXT PHASE PRIORITIES

    Phase 2: Complete Agent Registration

  • MemoryAgent & EnhancedMemoryAgent - Memory management
  • SimpleCoderAgent - Development capabilities
  • MultiModelAgent - LLM optimization
  • TestAgent & ReportAgent - Testing infrastructure

    • Phase 3: Advanced Security

  • Token-based access control
  • Role-based permissions
  • Cross-agent authentication
  • Blockchain integration readiness

    • 🏆 IDENTITY MANAGEMENT ACHIEVEMENT

    The mindX system has successfully implemented enterprise-grade identity management with:

  • 100% Tool Security: All 17 tools cryptographically secured
  • Critical Agent Registration: All 3 core agents registered
  • Enhanced Guardian Workflow: Registry validation integrated
  • Production-Ready Security: Challenge-response authentication
  • Scalable Architecture: Ready for additional agent registration

    • This represents a MAJOR MILESTONE in the mindX orchestration environment's evolution toward a fully autonomous, secure, and trusted multi-agent system.

    This documentation reflects the current state after the IDENTITY MANAGEMENT OVERHAUL and highlights the significant security improvements achieved.*

    All DocumentsDocument IndexThe Book of mindXImprovement JournalAPI Reference