agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4¶

Self-Discover Agent Implementation following LangGraph tutorial pattern.

Based on the official LangGraph Self-Discover tutorial: https://langchain-ai.github.io/langgraph/tutorials/self-discover/self-discover/

This implementation follows the exact pattern from the tutorial with proper state management and structured output parsing.

Classes¶

`AdaptedModulesOutput`	Output from module adapter - string format.
`FinalAnswerOutput`	Output from final reasoner - string format.
`ModuleSelectionOutput`	Output from module selector - string format.
`ReasoningStructureOutput`	Output from structure creator - string format.
`SelfDiscoverAdapter`	Agent that adapts modules to be task-specific.
`SelfDiscoverExecutor`	Agent that executes the reasoning plan.
`SelfDiscoverSelector`	Agent that selects relevant reasoning modules.
`SelfDiscoverState`	State for Self-Discover workflow following LangGraph tutorial.
`SelfDiscoverStructurer`	Agent that creates a structured reasoning plan.

Functions¶

`main`()	Example of using Self-Discover Enhanced V4.
`run_self_discover_workflow`(task[, modules])	Run the Self-Discover workflow sequentially.

Module Contents¶

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.AdaptedModulesOutput(/, **data)¶

Bases: pydantic.BaseModel

Output from module adapter - string format.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.FinalAnswerOutput(/, **data)¶

Bases: pydantic.BaseModel

Output from final reasoner - string format.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.ModuleSelectionOutput(/, **data)¶

Bases: pydantic.BaseModel

Output from module selector - string format.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.ReasoningStructureOutput(/, **data)¶

Bases: pydantic.BaseModel

Output from structure creator - string format.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.SelfDiscoverAdapter¶

Bases: SimpleAgentV3

Agent that adapts modules to be task-specific.

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.SelfDiscoverExecutor¶

Bases: SimpleAgentV3

Agent that executes the reasoning plan.

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.SelfDiscoverSelector¶

Bases: SimpleAgentV3

Agent that selects relevant reasoning modules.

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.SelfDiscoverState¶

Bases: TypedDict

State for Self-Discover workflow following LangGraph tutorial.

Initialize self. See help(type(self)) for accurate signature.

class agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.SelfDiscoverStructurer¶

Bases: SimpleAgentV3

Agent that creates a structured reasoning plan.

async agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.main()¶: Example of using Self-Discover Enhanced V4.

async agents.reasoning_and_critique.self_discover.self_discover_enhanced_v4.run_self_discover_workflow(task, modules=None)¶

Run the Self-Discover workflow sequentially.

Parameters:

task (str) – The task to solve
modules (str | None) – Optional custom reasoning modules

Returns:

Dict containing the final answer and reasoning

Return type:

dict[str, Any]