Skip to main content

​
Agent

Defined in the agent module.

​
Constructor

​
name
Optional
No description available.
​
role
Optional
No description available.
​
goal
Optional
No description available.
​
backstory
Optional
No description available.
​
instructions
Optional
No description available.
​
llm
Optional
No description available.
​
model
Optional
No description available.
​
function_calling_llm
Optional
No description available.
​
llm_config
Optional
No description available.
​
base_url
Optional
No description available.
​
api_key
Optional
No description available.
​
tools
Optional
No description available.
​
allow_delegation
bool
default:"False"
No description available.
​
allow_code_execution
Optional
default:"False"
No description available.
​
code_execution_mode
Literal
default:"'safe'"
No description available.
​
handoffs
Optional
No description available.
​
auto_save
Optional
No description available.
​
rate_limiter
Optional
No description available.
​
memory
Optional
No description available.
​
knowledge
Optional
No description available.
​
planning
Optional
default:"False"
No description available.
​
reflection
Optional
No description available.
​
guardrails
Optional
No description available.
​
web
Optional
No description available.
​
context
Optional
No description available.
​
autonomy
Optional
No description available.
​
verification_hooks
Optional
No description available.
​
output
Optional
No description available.
​
execution
Optional
No description available.
​
templates
Optional
No description available.
​
caching
Optional
No description available.
​
hooks
Optional
No description available.
​
skills
Optional
No description available.

​
Methods

from_template()

Create an Agent from a template.

stream_emitter()

Lazy-loaded StreamEventEmitter for real-time events (zero overhead when not used).

stream_emitter()

Allow setting stream_emitter directly.

auto_memory()

AutoMemory instance for automatic memory extraction.

auto_memory()

Instance method.

policy()

PolicyEngine instance for execution control.

policy()

Instance method.

background()

BackgroundRunner instance for async task execution.

background()

Instance method.

checkpoints()

CheckpointService instance for file-level undo/restore.

checkpoints()

Instance method.

output_style()

OutputStyle instance for response formatting.

output_style()

Instance method.

thinking_budget()

ThinkingBudget instance for extended thinking control.

thinking_budget()

Instance method.

context_manager()

ContextManager instance for unified context management.

context_manager()

Set context manager directly.

console()

Lazily initialize Rich Console only when needed AND verbose is True.

skill_manager()

Lazily initialize SkillManager only when skills are accessed.

get_skills_prompt()

Get the XML prompt for available skills.

agent_id()

Lazily generate agent ID when first accessed.

display_name()

Safe display name that never returns None.

analyze_prompt()

Analyze prompt for autonomy signals.

get_recommended_stage()

Get recommended execution stage for prompt.

run_autonomous()

Run an autonomous task execution loop.

handoff_to()

Programmatically hand off a task to another agent.

handoff_to_async()

Asynchronously hand off a task to another agent.

get_available_tools()

Get tools available to this agent, filtered by plan_mode if enabled.

rules_manager()

Lazy-initialized RulesManager for persistent rules/instructions.

get_rules_context()

Get rules context for the current conversation.

get_memory_context()

Get memory context for the current conversation.

store_memory()

Store content in memory.

llm_model()

Unified property to get the LLM model regardless of configuration type.

retrieval_config()

Get the unified retrieval configuration.

rag()

Lazy-loaded RAG instance for advanced retrieval with citations.

retrieve()

Retrieve context from knowledge without LLM generation.

query()

Query knowledge and get a structured answer with citations.

rag_query()

Query knowledge using RAG pipeline with citations.

chat_with_context()

Chat with pre-retrieved context.

generate_task()

Generate a Task object from the agent’s instructions

execute_tool()

Execute a tool dynamically based on the function name and arguments.

clear_history()

Instance method.

prune_history()

Prune chat history to keep only the last N messages.

delete_history()

Delete a specific message from chat history by index.

delete_history_matching()

Delete all messages matching a pattern.

get_history_size()

Get the current number of messages in chat history.

ephemeral()

Context manager for ephemeral conversations.

session_id()

Get the current session ID.

chat()

Chat with the agent.

clean_json_output()

Clean and extract JSON from response text.

achat()

Async version of chat method with self-reflection support.

arun()

Async version of run() - silent, non-streaming, returns structured result.

astart()

Async version of start() - interactive, streaming-aware.

run()

Execute agent silently and return structured result.

switch_model()

Switch the agent’s LLM model while preserving conversation history.

start()

Start the agent interactively with verbose output.

iter_stream()

Stream agent response as an iterator of chunks.

execute()

Execute a task synchronously - backward compatibility method

aexecute()

Execute a task asynchronously - backward compatibility method

execute_tool_async()

Async version of execute_tool

launch()

Launch the agent as an HTTP API endpoint or an MCP server.