Skip to main content

Interface: IPlanningEngine

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:393

Interface for the AgentOS Planning Engine.

The Planning Engine provides sophisticated cognitive capabilities for autonomous agents, including:

  • Goal Decomposition: Break complex goals into manageable subtasks
  • Plan Generation: Create multi-step execution plans with ReAct pattern
  • Self-Correction: Refine plans based on execution feedback
  • Autonomous Loops: Pursue goals with minimal human intervention

Example​

// Generate and execute a plan
const engine = new PlanningEngine(llmProvider);
const plan = await engine.generatePlan('Build a REST API', {
  strategy: 'plan_and_execute',
  maxSteps: 20,
});

for (const step of plan.steps) {
  const result = await engine.executeStep(step);
  if (!result.success) {
    const refined = await engine.refinePlan(plan, {
      planId: plan.planId,
      stepId: step.stepId,
      feedbackType: 'step_failed',
      details: result.error!,
      severity: 'error',
    });
  }
}

Methods​

decomposeTask()​

decomposeTask(task, depth?): Promise<TaskDecomposition>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:428

Decomposes a complex task into simpler subtasks.

Parameters​

task​

string

The task description to decompose

depth?​

number

Maximum decomposition depth (default: 3)

Returns​

Promise<TaskDecomposition>

Task decomposition result

executeStep()​

executeStep(step, context?): Promise<PlanStepResult>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:472

Executes a single plan step.

Parameters​

step​

PlanStep

Step to execute

context?​

StepExecutionContext

Execution context including previous results

Returns​

Promise<PlanStepResult>

Step execution result

generatePlan()​

generatePlan(goal, context?, options?): Promise<ExecutionPlan>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:415

Generates a multi-step execution plan from a high-level goal.

Parameters​

goal​

string

The high-level goal to achieve

context?​

PlanningContext

Additional context for planning

options?​

PlanningOptions

Planning configuration options

Returns​

Promise<ExecutionPlan>

Generated execution plan

Example​

const plan = await engine.generatePlan(
  'Analyze customer feedback and generate a report',
  { domainContext: 'E-commerce platform' },
  { strategy: 'react', maxSteps: 15 }
);

getExecutionState()​

getExecutionState(planId): ExecutionState | null

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:515

Gets the current execution state for a plan.

Parameters​

planId​

string

Plan identifier

Returns​

ExecutionState | null

Current execution state or null if not found

refinePlan()​

refinePlan(plan, feedback): Promise<ExecutionPlan>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:450

Refines an existing plan based on execution feedback. Uses self-reflection to identify issues and generate corrections.

Parameters​

plan​

ExecutionPlan

Original plan to refine

feedback​

ExecutionFeedback

Feedback from execution

Returns​

Promise<ExecutionPlan>

Refined execution plan

reflect()​

reflect(plan, executionState): Promise<ReflectionResult>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:459

Performs self-reflection on plan execution state.

Parameters​

plan​

ExecutionPlan

Current plan

executionState​

ExecutionState

Current execution state

Returns​

Promise<ReflectionResult>

Reflection insights and suggested adjustments

restoreCheckpoint()​

restoreCheckpoint(checkpointId): Promise<{ plan: ExecutionPlan; state: ExecutionState; }>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:507

Restores execution state from a checkpoint.

Parameters​

checkpointId​

string

Checkpoint to restore

Returns​

Promise<{ plan: ExecutionPlan; state: ExecutionState; }>

Restored execution state

runAutonomousLoop()​

runAutonomousLoop(goal, options?): AsyncGenerator<LoopProgress, ExecutionSummary, undefined>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:483

Runs an autonomous goal pursuit loop. Yields progress updates and handles self-correction automatically.

Parameters​

goal​

string

Goal to pursue

options?​

AutonomousLoopOptions

Loop configuration

Returns​

AsyncGenerator<LoopProgress, ExecutionSummary, undefined>

Final execution summary

Yields​

Progress updates including current step and observations

saveCheckpoint()​

saveCheckpoint(plan, state): Promise<string>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:499

Saves current execution state for checkpointing/rollback.

Parameters​

plan​

ExecutionPlan

Plan being executed

state​

ExecutionState

Current execution state

Returns​

Promise<string>

Checkpoint identifier

validatePlan()​

validatePlan(plan): Promise<PlanValidationResult>

Defined in: packages/agentos/src/core/planning/IPlanningEngine.ts:436

Validates a plan for feasibility and completeness.

Parameters​

plan​

ExecutionPlan

Plan to validate

Returns​

Promise<PlanValidationResult>

Validation result with any issues found