calute.cortex.core.enums

Enumerations for Cortex framework.

This module provides enumeration types used throughout the Cortex multi-agent orchestration framework. These enums define: - Process execution types for agent orchestration strategies - Chain types for task dependency structures

These enumerations enable type-safe configuration of agent workflows and ensure consistent behavior across the Cortex framework.

Example

>>> from calute.cortex.enums import ProcessType, ChainType
>>> process = ProcessType.SEQUENTIAL
>>> chain = ChainType.LINEAR

class calute.cortex.core.enums.ChainType(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)

Bases: Enum

Enumeration of chain types for task dependency structures.

Defines how tasks are connected and depend on each other within a workflow. Chain types determine the flow and structure of task execution in complex multi-step processes.

LINEAR

Tasks form a straight sequence with single dependencies. Each task depends on exactly one predecessor.

BRANCHING

Tasks can split into multiple parallel paths. One task can lead to multiple dependent tasks executing in parallel.

LOOP

Tasks can form cycles for iterative processing. Allows repetition of task sequences until a condition is met.

BRANCHING = 'branching'

LINEAR = 'linear'

LOOP = 'loop'

class calute.cortex.core.enums.ProcessType(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)

Bases: Enum

Enumeration of execution process types for Cortex orchestration.

Defines the available orchestration strategies for multi-agent task execution within the Cortex framework. Each process type determines how agents coordinate and execute tasks.

SEQUENTIAL

Tasks execute one after another in order. Each task waits for the previous one to complete before starting.

HIERARCHICAL

A manager agent delegates tasks to worker agents. The manager coordinates assignments and reviews outputs.

PARALLEL

Multiple tasks execute simultaneously. Independent tasks can run concurrently for faster completion.

CONSENSUS

Multiple agents work on the same task. Outputs are synthesized into a unified consensus response.

PLANNED

Tasks follow an AI-generated execution plan. A planner agent creates a detailed step-by-step workflow.

CONSENSUS = 'consensus'

HIERARCHICAL = 'hierarchical'

PARALLEL = 'parallel'

PLANNED = 'planned'

SEQUENTIAL = 'sequential'