API Reference¶

Complete API reference for AgenticAI Framework v2.0.

Core Classes¶

Agent¶

Python
from agenticaiframework import Agent

Constructor

Parameter	Type	Default	Description
`name`	`str`	required	Agent name
`role`	`str`	required	Role description
`capabilities`	`list[str]`	required	Agent capabilities
`config`	`dict[str, Any]`	required	Configuration dictionary
`max_context_tokens`	`int`	`4096`	Maximum context window tokens

Factory Methods

Method	Description
`Agent.quick(name, *, role, llm, provider, tools, auto_tools, guardrails, tracing)`	Create with sensible defaults
`Agent.from_config(config_dict)`	Create from configuration dictionary

Instance Methods

Method	Returns	Description
`execute(task)`	`AgentResult`	Execute a task
`invoke(prompt)`	`str`	Quick invocation with a prompt string

Role Templates: assistant, analyst, coder, writer, researcher

Task¶

Python
from agenticaiframework import Task

Parameter	Type	Default	Description
`description`	`str`	required	Task description
`agent`	`Agent`	`None`	Assigned agent
`expected_output`	`str`	`None`	Expected output description
`priority`	`int`	`0`	Task priority (higher = more important)

Process¶

Python
from agenticaiframework import Process

Parameter	Type	Default	Description
`name`	`str`	required	Process name
`strategy`	`str`	`"sequential"`	`"sequential"`, `"parallel"`, `"hybrid"`
`max_workers`	`int` or `None`	`None`	Thread pool size

Methods

Method	Returns	Description
`add_task(callable, args, *kwargs)`	`None`	Add a task
`add_step(callable, args, *kwargs)`	`None`	Alias for `add_task`
`execute()`	`list[Any]`	Run all tasks

Attributes (via __slots__): name, strategy, tasks, status, max_workers

Hub¶

Python
from agenticaiframework import Hub

Method	Returns	Description
`register(category, name, item)`	`None`	Register a component
`get(category, name)`	`Any`	Retrieve a component
`list_items(category)`	`dict`	List all items in a category
`remove(category, name)`	`None`	Remove a component
`register_service(name, service)`	`None`	Register a service
`get_service(name)`	`Any`	Retrieve a service

Valid categories: agents, prompts, tools, guardrails, llms, services

MonitoringSystem¶

Python
from agenticaiframework import MonitoringSystem

Method	Returns	Description
`record_metric(name, value)`	`None`	Record a numeric metric
`get_metric(name)`	`list`	Get values for a metric
`get_metrics()`	`dict`	Get all metrics
`log_event(event_type, details)`	`None`	Log a structured event
`get_events()`	`list[dict]`	Get all events
`log_message(message)`	`None`	Log a message
`get_logs()`	`list[str]`	Get all log messages
`get_gc_stats()`	`dict`	Get GC statistics
`force_gc()`	`int`	Force garbage collection
`clear()`	`None`	Reset all data

KnowledgeRetriever¶

Python
from agenticaiframework import KnowledgeRetriever

Method	Returns	Description
`register_source(name, retrieval_fn)`	`None`	Register a retrieval source
`add_knowledge(key, content)`	`None`	Add static knowledge
`retrieve(source, query, use_cache=True)`	`Any`	Query a source
`clear_cache()`	`None`	Clear the LRU cache

Workflow¶

Python
from agenticaiframework import Workflow

Method	Returns	Description
`add_step(name, callable, args, *kwargs)`	`None`	Add a workflow step
`run()`	`list[Any]`	Execute the workflow

Configurations¶

Python
from agenticaiframework import Configurations

Method	Returns	Description
`set(key, value)`	`None`	Set a configuration value
`get(key, default=None)`	`Any`	Get a configuration value

MCP Tools¶

MCPTool¶

Python
from agenticaiframework.mcp_tools import MCPTool

Attribute	Type	Description
`id`	`str`	Unique tool identifier
`name`	`str`	Human-readable name
`capability`	`str`	Tool description
`execute_fn`	`Callable`	Function to invoke

Uses __slots__.

MCPToolManager¶

Python
from agenticaiframework.mcp_tools import MCPToolManager

Method	Returns	Description
`register_tool(tool)`	`None`	Register an MCPTool
`execute_tool(tool_id, **kwargs)`	`Any`	Execute by ID
`execute_tool_by_name(name, **kwargs)`	`Any`	Execute by name
`list_tools()`	`list[MCPTool]`	List all tools
`remove_tool(tool_id)`	`None`	Remove a tool

Orchestration¶

AgentTeam¶

Python
from agenticaiframework.orchestration import AgentTeam

Method	Returns	Description
`add_member(agent)`	`None`	Add an agent to the team
`remove_member(agent_id)`	`None`	Remove an agent

AgentSupervisor¶

Python
from agenticaiframework.orchestration import AgentSupervisor

Parameter	Type	Description
`name`	`str`	Supervisor name
`agents`	`list[Agent]`	Managed agents

Security¶

InputValidator¶

Python
from agenticaiframework.security import InputValidator

Method	Returns	Description
`validate(text)`	`tuple[bool, str]`	Validate and sanitise input

PIIMasker¶

Python
from agenticaiframework.compliance import PIIMasker

Method	Returns	Description
`mask(text)`	`str`	Mask PII in text

Tracing¶

TracingManager¶

Python
from agenticaiframework.tracing import TracingManager

Method	Returns	Description
`span(name)`	context manager	Create a tracing span

Memory¶

MemoryManager¶

Python
from agenticaiframework.memory import MemoryManager

Method	Returns	Description
`store(content, metadata)`	`None`	Store a memory
`search(query, top_k)`	`list`	Search memories

Exceptions¶

Python
from agenticaiframework.exceptions import (
    AgenticAIError, # Base exception
    ConfigurationError, # Configuration issues
    ProcessExecutionError, # Process execution failures
    AgentError, # Agent-related errors
    AgentNotFoundError, # Agent not found
    AgentExecutionError, # Agent execution failures
)

Quick Start — get started in 5 minutes
Usage Guide — everyday patterns
Configuration — framework configuration
Best Practices — production patterns