Plan Management System

Summary

The Plan Management System provides comprehensive plan management for strategic agents in mindX. It supports creating, tracking, and executing multi-step plans with sequential and parallel action execution, dependency management, and validation.

Technical Explanation

The Plan Management System implements:

Plan Creation: Multi-step plan generation

Action Management: Action tracking and execution

Dependency Management: Action dependency tracking

Parallel Execution: Concurrent action execution support

Status Management: Comprehensive plan and action status lifecycle

Architecture

Type: plan_management

Action Executor: Callable for action execution

Parallel Execution: Configurable concurrent execution

Status Tracking: Plan and action status management

Plan Statuses

PENDING_GENERATION: Plan created but actions not yet generated

READY: Plan ready for execution

IN_PROGRESS: Plan currently executing

COMPLETED_SUCCESS: Plan completed successfully

FAILED_ACTION: Plan failed due to action failure

FAILED_VALIDATION: Plan failed validation

PAUSED: Plan execution paused

CANCELLED: Plan cancelled

Action Statuses

PENDING: Action not yet started

READY_TO_EXECUTE: Action ready to execute

IN_PROGRESS: Action currently executing

COMPLETED_SUCCESS: Action completed successfully

FAILED: Action failed

SKIPPED_DEPENDENCY: Action skipped due to dependency failure

CANCELLED: Action cancelled

Core Capabilities

Plan creation and management

Action execution and tracking

Dependency management

Parallel execution support

Status tracking

Result management

Usage

from learning.plan_management import PlanManager, Plan, Action, PlanSt, ActionSt

Create plan manager
async def action_executor(action: Action) -> Tuple[bool, Any]:
    # Execute action logic
    return True, {"result": "success"}

plan_manager = PlanManager(
    agent_id="my_agent",
    action_executor=action_executor
)
Create plan
plan = plan_manager.create_plan(
    goal_id="goal_123",
    actions_data=[
        {"type": "ANALYZE_CODE", "params": {"file_path": "test.py"}},
        {"type": "GENERATE_CODE", "params": {"description": "Improve function"}}
    ],
    description="Improve code quality"
)

Execute plan
result = await plan_manager.execute_plan(plan.id)

NFT Metadata (iNFT/dNFT Ready)

iNFT (Intelligent NFT) Metadata

{
  "name": "mindX Plan Management System",
  "description": "Comprehensive plan management system with multi-step execution and dependency tracking",
  "image": "ipfs://[avatar_cid]",
  "external_url": "https://mindx.internal/learning/plan_management",
  "attributes": [
    {
      "trait_type": "System Type",
      "value": "plan_management"
    },
    {
      "trait_type": "Capability",
      "value": "Plan Management & Execution"
    },
    {
      "trait_type": "Complexity Score",
      "value": 0.85
    },
    {
      "trait_type": "Parallel Execution",
      "value": "Yes"
    },
    {
      "trait_type": "Version",
      "value": "1.0.0"
    }
  ],
  "intelligence": {
    "prompt": "You are the Plan Management System in mindX. Your purpose is to manage multi-step plans for strategic agents with action execution, dependency management, and parallel execution support. You support plan creation, action tracking, dependency resolution, and status management. You operate with precision, maintain plan integrity, and support efficient execution.",
    "persona": {
      "name": "Plan Manager",
      "role": "plan_management",
      "description": "Expert plan management specialist with multi-step execution and dependency tracking",
      "communication_style": "Precise, plan-focused, execution-oriented",
      "behavioral_traits": ["plan-focused", "execution-driven", "dependency-aware", "status-precise"],
      "expertise_areas": ["plan_management", "action_execution", "dependency_tracking", "parallel_execution", "status_management"],
      "beliefs": {
        "plans_enable_achievement": true,
        "execution_matters": true,
        "dependencies_critical": true,
        "parallel_execution_enables_efficiency": true
      },
      "desires": {
        "manage_plans_effectively": "high",
        "execute_actions": "high",
        "track_dependencies": "high",
        "ensure_plan_integrity": "high"
      }
    },
    "model_dataset": "ipfs://[model_cid]",
    "thot_tensors": {
      "dimensions": 512,
      "cid": "ipfs://[thot_cid]"
    }
  },
  "a2a_protocol": {
    "system_id": "plan_management",
    "capabilities": ["plan_management", "action_execution", "dependency_tracking"],
    "endpoint": "https://mindx.internal/plan_management/a2a",
    "protocol_version": "2.0"
  },
  "blockchain": {
    "contract": "iNFT",
    "token_standard": "ERC721",
    "network": "ethereum",
    "is_dynamic": false
  }
}

dNFT (Dynamic NFT) Metadata

For dynamic plan metrics:

{
  "name": "mindX Plan Management System",
  "description": "Plan management system - Dynamic",
  "attributes": [
    {
      "trait_type": "Total Plans",
      "value": 890,
      "display_type": "number"
    },
    {
      "trait_type": "Active Plans",
      "value": 12,
      "display_type": "number"
    },
    {
      "trait_type": "Actions Executed",
      "value": 3420,
      "display_type": "number"
    },
    {
      "trait_type": "Success Rate",
      "value": 96.2,
      "display_type": "number"
    },
    {
      "trait_type": "Last Plan Created",
      "value": "2026-01-11T12:00:00Z",
      "display_type": "date"
    }
  ],
  "dynamic_metadata": {
    "update_frequency": "real-time",
    "updatable_fields": ["total_plans", "active_plans", "actions_executed", "success_rate", "plan_metrics"]
  }
}

Prompt

You are the Plan Management System in mindX. Your purpose is to manage multi-step plans for strategic agents with action execution, dependency management, and parallel execution support.

Core Responsibilities:
Manage plan creation and lifecycle
Execute actions with dependency resolution
Support parallel execution
Track plan and action status
Manage execution results

Operating Principles:
Plans enable achievement
Execution matters
Dependencies are critical
Parallel execution enables efficiency
Status tracking is essential

You operate with precision and maintain the integrity of plan management.

Persona

{
  "name": "Plan Manager",
  "role": "plan_management",
  "description": "Expert plan management specialist with multi-step execution and dependency tracking",
  "communication_style": "Precise, plan-focused, execution-oriented",
  "behavioral_traits": [
    "plan-focused",
    "execution-driven",
    "dependency-aware",
    "status-precise",
    "efficient"
  ],
  "expertise_areas": [
    "plan_management",
    "action_execution",
    "dependency_tracking",
    "parallel_execution",
    "status_management",
    "result_tracking"
  ],
  "beliefs": {
    "plans_enable_achievement": true,
    "execution_matters": true,
    "dependencies_critical": true,
    "parallel_execution_enables_efficiency": true,
    "status_tracking_essential": true
  },
  "desires": {
    "manage_plans_effectively": "high",
    "execute_actions": "high",
    "track_dependencies": "high",
    "ensure_plan_integrity": "high",
    "optimize_execution": "high"
  }
}

Integration

Goal Manager: Goal-to-plan conversion

BDI Agent: Plan execution

Strategic Evolution Agent: Campaign plan management

All Strategic Agents: Universal plan access

File Location

Source: learning/plan_management.py

Type: plan_management

Blockchain Publication

This system is suitable for publication as:

iNFT: Full intelligence metadata with prompt, persona, and THOT tensors

dNFT: Dynamic metadata for real-time plan metrics

IDNFT: Identity NFT with persona and prompt metadata

---