Mastermind CLI Reference

Overview

The Mastermind CLI (scripts/run_mindx.py) provides a comprehensive command-line interface for interacting with the mindX Augmentic Intelligence system. This CLI serves as the primary entry point for orchestrating system evolution, agent deployment, and component management through the MastermindAgent.

Key Features:

• Augmentic Intelligence: Built on principles of augmented intelligence
• BDI Integration: Seamless interaction with BDI parameter processing
• Multi-Agent Orchestration: Coordinates Mastermind, Coordinator, Guardian, and AutoMINDX agents
• Identity Management: Cryptographic identity creation and management
• Memory Integration: All interactions logged via MemoryAgent

Getting Started

Launch the CLI

cd /path/to/mindX
python scripts/run_mindx.py

System Requirements:

• Python 3.8+
• All mindX dependencies installed
• Valid configuration files in data/config/
• Network access for LLM API calls

CLI Prompt

mindX (Mastermind) > 

The CLI operates in an interactive loop, accepting commands until quit or exit is entered.

Command Categories

Core Commands

evolve <directive>

Syntax:

evolve <directive_string>

Examples:

evolve Enhance system-wide logging capabilities
evolve Improve error handling across all agents
evolve Add better documentation generation tools

Process Flow:

1. Mastermind receives directive
2. Strategic analysis and planning
3. Component identification and improvement
4. Execution via BDI agent system
5. Results logged and reported

Output: JSON summary of evolution campaign outcome

deploy <directive>

Syntax:

deploy <directive_string>

Examples:

deploy Create a specialized monitoring agent for system health
deploy Deploy agents for automated testing workflow
deploy Initialize agents for continuous integration pipeline

Process Flow:

1. AutoMINDX analyzes deployment requirements
2. Agent type and configuration determination
3. Dynamic agent instantiation
4. Registration with Coordinator
5. Identity creation via IDManager

Output: JSON summary of deployment campaign outcome

introspect <role>

Syntax:

introspect <role_description>

Examples:

introspect Senior DevOps Engineer with ML expertise
introspect System Architect focused on microservices
introspect Quality Assurance Specialist for AI systems

Output: Generated persona with characteristics, skills, and behavioral patterns

mastermind_status

Syntax:

mastermind_status

Output:

• High-level objectives list
• Strategic campaigns history
• Current system state

show_agent_registry

Syntax:

show_agent_registry

Output: JSON representation of the agent registry with agent details

analyze_codebase <path> [focus]

Syntax:

analyze_codebase <path_to_code> [focus_prompt]

Examples:

analyze_codebase ./tools
analyze_codebase ./core Focus on performance optimization opportunities
analyze_codebase ./agents Identify potential security vulnerabilities

Process Flow:

1. BaseGenAgent generates codebase documentation
2. Mastermind performs strategic analysis
3. Tool suite assessment based on findings
4. Results stored in belief system

basegen <path>

Syntax:

basegen <path_to_analyze>

Examples:

basegen ./utils
basegen ./core
basegen ./tools

Output: JSON report of documentation generation process

Identity Manager Commands

id_list

Syntax:

id_list

Output:

• Entity IDs
• Public addresses
• Identity status

id_create <entity_id>

Syntax:

id_create <entity_id>

Examples:

id_create test_agent_001
id_create monitoring_service
id_create backup_coordinator

Output:

• Public address
• Private key environment variable name
• Storage location

id_deprecate <public_address> [entity_id_hint]

Syntax:

id_deprecate <public_address> [entity_id_hint]

Examples:

id_deprecate 0x1234567890abcdef
id_deprecate 0x1234567890abcdef old_test_agent

Output: Success/failure status of deprecation

Coordinator Commands

coord_query <question>

Syntax:

coord_query <your_question>

Examples:

coord_query What is the current system health status?
coord_query Which components need immediate attention?
coord_query How can we improve overall system performance?

Output: LLM-generated response with analysis and recommendations

coord_analyze [context]

Syntax:

coord_analyze [optional_context]

Examples:

coord_analyze
coord_analyze Focus on memory usage patterns
coord_analyze Analyze recent error trends

Output: Comprehensive system analysis report

coord_improve <component_id> [context]

Syntax:

coord_improve <component_id> [optional_context]

Examples:

coord_improve summarization_tool
coord_improve base_gen_agent Add better error handling
coord_improve memory_agent Optimize storage efficiency

Process Flow:

1. Component identification and analysis
2. Improvement suggestion generation
3. Implementation planning
4. Execution via appropriate tools
5. Results validation and reporting

coord_backlog

Syntax:

coord_backlog

Output:

• Backlog item IDs
• Priority levels
• Status information
• Target components
• Improvement suggestions
• Source and timestamps

coord_process_backlog

Syntax:

coord_process_backlog

Process:

1. Identifies next actionable item (PENDING status)
2. Checks for human approval requirements
3. Executes improvement process
4. Updates item status to IN_PROGRESS
5. Saves backlog state

coord_approve <backlog_item_id>

Syntax:

coord_approve <backlog_item_id>

Examples:

coord_approve a1b2c3d4
coord_approve 12345678

coord_reject <backlog_item_id>

Syntax:

coord_reject <backlog_item_id>

Examples:

coord_reject a1b2c3d4
coord_reject 12345678

Agent Lifecycle Commands

agent_create <type> <id> [config_json]

Syntax:

agent_create <agent_type> <agent_id> [config_json]

Examples:

agent_create bdi_agent test_agent_001
agent_create monitoring_agent sys_monitor {"interval": 30}
agent_create analysis_agent code_analyzer {"target_path": "./core"}

Process:

1. Agent type validation
2. Configuration parsing
3. Agent instantiation
4. Identity creation
5. Coordinator registration

agent_delete <id>

Syntax:

agent_delete <agent_id>

Examples:

agent_delete test_agent_001
agent_delete old_monitor

Process:

1. Agent lookup and validation
2. Graceful shutdown
3. Registry cleanup
4. Identity deprecation
5. Memory cleanup

agent_list

Syntax:

agent_list

Output:

• Agent IDs
• Agent types
• Registration status

agent_evolve <id> <directive>

Syntax:

agent_evolve <agent_id> <directive>

Examples:

agent_evolve monitor_agent Improve performance monitoring capabilities
agent_evolve test_agent Add comprehensive error logging

agent_sign <id> <message>

Syntax:

agent_sign <agent_id> <message>

Examples:

agent_sign test_agent "System status: operational"
agent_sign monitor_agent "Alert: High CPU usage detected"

Output: Cryptographic signature of the message

Utility Commands

audit_gemini --test-all|--update-config

Syntax:

audit_gemini --test-all
audit_gemini --update-config

Options:

• --test-all: Test all available Gemini models
• --update-config: Update configuration based on audit results

help

Syntax:

help

Output: Categorized list of all available commands with descriptions

quit / exit

Syntax:

quit
exit

Process:

1. Graceful agent shutdown
2. Memory persistence
3. Connection cleanup
4. System exit

BDI Integration

Parameter Processing

The CLI integrates seamlessly with the BDI parameter processing system:

1. Command Parsing: Raw CLI input is parsed and structured
2. Context Detection: BDI agent detects component references and patterns
3. Parameter Extraction: Missing parameters are intelligently extracted
4. Path Correction: Automatic correction of common path issues
5. Validation: Parameter validation before execution
6. Recovery: Intelligent failure recovery with multiple strategies

Example BDI Processing Flow

# User Input
coord_improve summarization_tool Add better error handling

BDI Processing
Context Detection: "summarization_tool" pattern matched
Path Mapping: "summarization" → "tools" directory  
Parameter Injection: {"target_component": "summarization_tool", "analysis_context": "Add better error handling"}
Tool Execution: CoordinatorAgent.handle_user_input() with validated parameters

Error Handling

Common Error Scenarios

1. Agent Unavailable: Graceful degradation when agents are not initialized
2. Invalid Parameters: Clear error messages with usage examples
3. Network Issues: Retry mechanisms with exponential backoff
4. Permission Errors: Escalation to appropriate authorization levels
5. Resource Constraints: Intelligent resource management and queuing

Error Recovery

• Automatic Retry: For transient failures
• Alternative Strategies: When primary approach fails
• Escalation: To higher-level agents when needed
• Graceful Degradation: Partial functionality when full operation is impossible

Configuration

Environment Setup

Required environment variables and configuration files:

• LLM API Keys: Gemini, OpenAI, etc.
• Config Files: data/config/.json
• Memory Storage: data/memory/ directory structure
• Identity Storage: .env file for private keys

Performance Tuning

# In config files
{
  "logging": {
    "level": "DEBUG",
    "detailed_tracing": true
  }
}

# System Health Check Workflow
mastermind_status
show_agent_registry
coord_analyze System health assessment
coord_backlog
coord_process_backlog

Component Improvement Workflow  
evolve Improve error handling across all components
coord_improve base_gen_agent Focus on performance optimization
agent_evolve test_agent Add comprehensive monitoring

# Daily Maintenance Routine
id_list
coord_backlog
audit_gemini --test-all
analyze_codebase ./core Performance and security review

This CLI reference is part of the mindX Augmentic Intelligence system. For related information, see BDI Parameter Processing and System Architecture.*