agents.planning.p_and_e.models¶

Models for Plan and Execute Agent System.

This module defines the data models for planning, execution, and replanning in the Plan and Execute agent architecture.

Classes¶

`Act`	Action to perform - either respond with answer or continue with plan.
`ExecutionResult`	Result from executing a single step.
`Plan`	Complete execution plan with steps and metadata.
`PlanStep`	Individual step in an execution plan.
`ReplanDecision`	Decision on whether to replan or provide final answer.
`Response`	Response to user with final answer.
`StepStatus`	Status of a plan step.
`StepType`	Type of plan step.

Module Contents¶

class agents.planning.p_and_e.models.Act(/, **data)¶

Bases: pydantic.BaseModel

Action to perform - either respond with answer or continue with plan.

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)

property is_final_response: bool¶

Check if this is a final response.

Return type:: bool

property is_plan: bool¶

Check if this is a plan to execute.

Return type:: bool

class agents.planning.p_and_e.models.ExecutionResult(/, **data)¶

Bases: pydantic.BaseModel

Result from executing a single step.

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)

to_prompt_format()¶

Format result for inclusion in prompts.

Return type:: str

class agents.planning.p_and_e.models.Plan(/, **data)¶

Bases: pydantic.BaseModel

Complete execution plan with steps and metadata.

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)

get_step(step_id)¶

Get a specific step by ID.

Parameters:: step_id (int)
Return type:: PlanStep | None

serialize_datetime(dt)¶

Serialize datetime fields to ISO format.

Parameters:: dt (datetime.datetime)
Return type:: str

to_prompt_format()¶

Format plan for inclusion in prompts.

Return type:: str

update_step_status(step_id, status, result=None, error=None)¶

Update the status of a specific step.

Parameters:

step_id (int)
status (StepStatus)
result (str | None)
error (str | None)

Return type:

bool

update_total_steps()¶

Ensure total_steps matches actual step count.

Return type:: Plan

classmethod validate_dependencies(steps)¶

Ensure dependencies reference valid step IDs.

Parameters:: steps (list[PlanStep])
Return type:: list[PlanStep]

classmethod validate_step_ids(steps)¶

Ensure step IDs are sequential starting from 1.

Parameters:: steps (list[PlanStep])
Return type:: list[PlanStep]

property completed_steps: list[PlanStep]¶

Get all completed steps.

Return type:: list[PlanStep]

property failed_steps: list[PlanStep]¶

Get all failed steps.

Return type:: list[PlanStep]

property has_failures: bool¶

Check if any steps have failed.

Return type:: bool

property is_complete: bool¶

Check if all steps are completed.

Return type:: bool

model_config¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

property next_step: PlanStep | None¶

Get the next step ready for execution.

Return type:: PlanStep | None

property pending_steps: list[PlanStep]¶

Get all pending steps.

Return type:: list[PlanStep]

property progress_percentage: float¶

Calculate completion percentage.

Return type:: float

class agents.planning.p_and_e.models.PlanStep(/, **data)¶

Bases: pydantic.BaseModel

Individual step in an execution plan.

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)

serialize_datetime(dt)¶

Serialize datetime fields to ISO format.

Parameters:: dt (datetime.datetime | None)
Return type:: str | None

to_prompt_format()¶

Format step for inclusion in prompts.

Return type:: str

property execution_time: float | None¶

Calculate execution time in seconds.

Return type:: float | None

property is_ready: bool¶

Check if step is ready to execute (all dependencies completed).

Return type:: bool

model_config¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class agents.planning.p_and_e.models.ReplanDecision(/, **data)¶

Bases: pydantic.BaseModel

Decision on whether to replan or provide final answer.

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)

validate_decision_fields()¶

Ensure required fields are present based on decision.

Return type:: ReplanDecision

class agents.planning.p_and_e.models.Response(/, **data)¶

Bases: pydantic.BaseModel

Response to user with final answer.

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.planning.p_and_e.models.StepStatus¶

Bases: str, enum.Enum

Status of a plan step.

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

class agents.planning.p_and_e.models.StepType¶

Bases: str, enum.Enum

Type of plan step.

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