ConditionalReplanningTool

All Implemented Interfaces:

com.embabel.agent.api.tool.DelegatingTool , com.embabel.agent.api.tool.Tool , com.embabel.agent.api.tool.ToolInfo
```
public final class ConditionalReplanningTool
 implements DelegatingTool
                    
```
Tool decorator that executes the wrapped tool, then conditionally triggers replanning based on the result.
Unlike ReplanningTool which always triggers replanning, this tool allows the ReplanDecider to inspect the result and decide whether to replan.
Note: This tool accesses AgentProcess via thread-local at call time, which is set by the decorator chain.

Nested Class Summary

Nested Classes
Modifier and Type	Class	Description

Field Summary

Fields
Modifier and Type	Field	Description
`private final Tool.Definition`	`definition`
`private final Tool.Metadata`	`metadata`
`private final Tool`	`delegate`

Constructor Summary

Constructors
Constructor	Description
`ConditionalReplanningTool(Tool delegate, ReplanDecider decider)`

Enum Constant Summary

Enum Constants
Enum Constant	Description

Method Summary

Modifier and Type	Method	Description
`Tool.Definition`	`getDefinition()`	Tool definition for LLM
`Tool.Metadata`	`getMetadata()`	Optional metadata
`Tool`	`getDelegate()`
`Tool.Result`	`call(String input)`	Execute the tool with JSON input.

Methods inherited from class com.embabel.agent.api.tool.Tool
withDescription, withName, withNote
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Constructor Detail
  - ConditionalReplanningTool
```
ConditionalReplanningTool(Tool delegate, ReplanDecider decider)
```
    Parameters:
    
    delegate - The tool to wrap
    
    decider - Decider that inspects the result context and determines whether to replan
- Method Detail
  - getDefinition
```
 Tool.Definition getDefinition()
```
    Tool definition for LLM
  - getMetadata
```
 Tool.Metadata getMetadata()
```
    Optional metadata
  - getDelegate
```
 Tool getDelegate()
```
  - call
```
 Tool.Result call(String input)
```
    Execute the tool with JSON input.
    
    Parameters:
    
    input - JSON string matching inputSchema
    
    Returns:
    
    Result to send back to LLM