Namespace LMKit.Agents.Planning
Classes
- ChainOfThoughtHandler
Planning handler that implements Chain-of-Thought (CoT) prompting.
Encourages the model to reason through problems step-by-step before providing a final answer. Improves performance on arithmetic, logic, and multi-step reasoning tasks.
- NonePlanningHandler
Planning handler that performs no explicit planning.
Passes input directly to the model and returns output unchanged. Use this for simple Q&A or when the model handles reasoning internally.
- PlanAndExecuteHandler
Planning handler that implements the Plan-and-Execute strategy.
First generates a complete plan for the task, then executes each step sequentially. Useful for complex multi-step tasks where having an upfront plan improves coherence.
- PlanningContext
Provides context and state for planning strategy execution.
The context is passed to IPlanningHandler methods and contains the agent configuration, execution state, and accumulated results.
- PlanningHandlerBase
Base class for planning handler implementations.
Provides common functionality and default implementations for planning handlers.
- PlanningHandlerFactory
Factory for creating IPlanningHandler instances based on strategy.
- PlanningStep
Represents a single step in the planning process.
Each step captures the thought, action, and observation from one iteration of the planning loop.
- PlanningStepResult
Represents the result of processing a planning step.
Returned by ProcessOutput(PlanningContext, string) to indicate the planning state and provide extracted content.
- ReActHandler
Planning handler that implements the ReAct (Reasoning + Acting) pattern.
Interleaves reasoning traces with tool actions in a Thought-Observation loop. Tool invocations are handled by the native tool-calling system of MultiTurnConversation, while this handler adds structured reasoning around the process.
- ReflectionHandler
Planning handler that implements reflection-based reasoning.
The agent generates an initial response, then critically evaluates and refines it through self-reflection. Improves output quality by catching errors and inconsistencies.
- TreeOfThoughtHandler
Planning handler that implements Tree-of-Thought (ToT) reasoning.
Explores multiple reasoning paths, evaluates intermediate states, and selects the most promising branches. Best for problems with multiple valid solution paths where exploration can find better solutions than linear reasoning.
Interfaces
- IPlanningHandler
Defines the contract for planning strategy implementations.
A planning handler transforms user input into a structured reasoning process, optionally modifying prompts, managing intermediate steps, and producing reasoning traces.
Enums
- PlanningStepStatus
Indicates the status after processing a planning step.
- PlanningStepType
Identifies the type of a planning step.