MindX Agent Documentation

Summary

The MindX Agent is the meta-agent that serves as the "execution mind" of the mindX Gödel machine. It understands all agents' roles, capabilities, and powers, and orchestrates them for continuous self-improvement of the mindX system itself. It is subservient to higher intelligence and acts as the sovereign intelligence that knows and improves the entire mindX system.

Technical Explanation

The MindX Agent implements:

• Meta-Awareness: Comprehensive understanding of all agents in the system
• Agent Knowledge Base: Maintains detailed knowledge of all agents' capabilities, roles, and powers
• Registry Integration: Uses Registry Manager Tool to track registered agents
• Identity Tracking: Uses ID Manager Agent to track agent identities
• Dynamic Agent Tracking: Monitors and tracks newly created agents from Agent Builder Agent
• Self-Improvement Orchestration: Uses SEA, BDI, Mastermind, and all other agents to improve mindX
• Memory Feedback: Gets context from Memory Agent and data/ folder
• Result Analysis: Compares actual results vs expected outcomes for continuous improvement
• Gödel Machine Execution: Can reason about and improve the system it's part of

Architecture

• Type: meta_agent
• Layer: Meta-layer above all agents
• Hierarchy: Higher Intelligence → mindXagent → All Other Agents
• Pattern: Meta-agent with comprehensive system understanding

Core Components

1. Agent Knowledge Base: Dictionary mapping agent_id to AgentKnowledge
2. Agent Capabilities Map: Detailed capability analysis for each agent
3. Agent Relationship Graph: Understanding of agent interactions
4. Registry Integration: Connection to Registry Manager Tool
5. Identity Tracking: Connection to ID Manager Agent
6. Agent Builder Integration: Monitors Agent Builder Agent for new agents
7. Memory Integration: Gets feedback from Memory Agent
8. Monitoring Integration: Uses monitoring agents for health tracking

Usage

from agents.core.mindXagent import MindXAgent

Get MindX Agent instance
mindx_agent = await MindXAgent.get_instance(
    agent_id="mindx_meta_agent",
    config=config,
    memory_agent=memory_agent,
    belief_system=belief_system
)

Build comprehensive agent knowledge base
all_agents = await mindx_agent.understand_all_agents()

Orchestrate self-improvement
result = await mindx_agent.orchestrate_self_improvement(
    "Improve system performance and reliability"
)

Get memory feedback
memory_context = await mindx_agent.get_memory_feedback("system improvement")

Monitor system health
health = await mindx_agent.monitor_system_health()

Analyze results
analysis = await mindx_agent.analyze_actual_results(task_id)

Key Methods

Agent Knowledge Management

• load_registered_agents(): Load all registered agents from registry
• track_agent_identities(): Track agent identities using ID Manager
• discover_agents_from_filesystem(): Discover agents by scanning filesystem
• understand_all_agents(): Build comprehensive knowledge base
• analyze_agent_capabilities(agent_id): Deep analysis of agent capabilities
• monitor_new_agents(): Monitor for newly created agents
• update_agent_knowledge(agent_id, new_capabilities): Update knowledge when agents evolve

Self-Improvement Orchestration

• orchestrate_self_improvement(improvement_goal): Orchestrate self-improvement using all agents
• execute_improvement_campaign(goal): Execute improvement campaign with result tracking
• select_agents_for_task(task): Intelligently select agents for tasks
• evolve_architecture(evolution_plan): Guide system architecture evolution

Memory and Feedback

• get_memory_feedback(context): Get feedback from Memory Agent and data/ folder
• analyze_actual_results(task_id): Analyze actual vs expected results
• monitor_system_health(): Monitor overall system health

Integration Points

Registry and Identity

• Registry Manager Tool: Tracks registered agents from official registry
• ID Manager Agent: Tracks agent identities and cryptographic keys
• Identity Tools: Identity sync and management

Agent Builder

• Agent Builder Agent: Monitors for newly created agents
• Dynamic Tracking: Automatically tracks new agents and their capabilities
• Event Notifications: Receives notifications when agents are created

Memory System

• Memory Agent: Gets context and feedback from memory system
• Data Folder: Monitors data/ folder for system state
• Improvement History: Tracks improvement history and lessons learned

Orchestration Agents

• StrategicEvolutionAgent: For improvement campaigns
• BDI Agent: For goal planning and cognitive reasoning
• Mastermind Agent: For strategic orchestration
• Coordinator Agent: For agent lifecycle and system services
• CEO Agent: For business strategy (when needed)

Monitoring Agents

• Performance Monitor: For performance metrics
• Resource Monitor: For resource metrics
• Error Recovery Coordinator: For error recovery

Self-Improvement Workflow

1. Goal Definition: Define improvement goal for mindX system
2. Agent Analysis: Analyze which agents are needed (including newly created agents)
3. Memory Context: Get context from Memory Agent and data/ folder
4. Campaign Creation: Use SEA to create improvement campaign
5. BDI Planning: Use BDI for detailed planning
6. Mastermind Coordination: Use Mastermind for strategic coordination
7. Execution: Coordinate execution through appropriate agents
8. Monitoring: Use monitoring agents to track progress and actual results
9. Result Analysis: Analyze actual results vs expected outcomes
10. Memory Feedback: Collect feedback from Memory Agent
11. Evaluation: Assess improvement results with real-world performance data
12. Learning: Update knowledge base with lessons learned
13. Continuous Improvement: Use feedback loop to continuously improve

Agent Knowledge Structure

Each agent in the knowledge base contains:

• agent_id: Unique identifier
• agent_type: Type (orchestration, core, learning, monitoring, specialized)
• location: File path
• capabilities: List of capabilities
• roles: List of roles
• powers: Dictionary of what it can do, limits, dependencies
• integration_points: Other agents it interacts with
• documentation: Documentation information
• status: Current status (ACTIVE, INACTIVE, etc.)
• identity: Cryptographic identity information
• registry_info: Registry information
• performance_metrics: Performance characteristics

Gödel Machine Aspects

• Self-Reference: Can reason about itself and the system it's part of
• Self-Modification: Can orchestrate changes to the system
• Meta-Learning: Learns about agents and system capabilities
• Recursive Improvement: Continuously improves the improvement process itself
• Dynamic Adaptation: Adapts as new agents are created

File Location

• Path: agents/core/mindXagent.py
• Documentation: docs/mindXagent.md

NFT Metadata

• Type: meta_agent
• Complexity: 0.99
• NFT Ready: ✅ iNFT, dNFT, IDNFT

Integration with Agent Builder Agent

The MindX Agent integrates with Agent Builder Agent to track newly created agents:

1. Agent Builder Agent creates new agent from prompt/request
2. Agent Builder Agent registers agent with Registry Manager Tool
3. Agent Builder Agent notifies MindX Agent
4. MindX Agent adds new agent to knowledge base
5. MindX Agent analyzes new agent's capabilities
6. MindX Agent updates agent relationship graph
7. MindX Agent monitors new agent's performance

Continuous Self-Improvement

The MindX Agent continuously improves mindX by:

1. Understanding all agents and their capabilities
2. Identifying improvement opportunities
3. Orchestrating improvement campaigns using appropriate agents
4. Monitoring actual results vs expected outcomes
5. Learning from memory feedback and results
6. Adapting improvement strategies based on what works
7. Tracking newly created agents and incorporating them into improvements

---

Last Updated: 2026-01-13 Status: ✅ Active Version: 1.0.0

---