Implementation:HKUDS AI Trader Ainvoke With Retry

Knowledge Sources	AI-Trader
Domains	Reliability, LLM_Agents
Last Updated	2026-02-09 14:00 GMT

Overview

Concrete tool for retrying LLM agent invocations with exponential backoff within the BaseAgent class.

Description

The BaseAgent._ainvoke_with_retry() async method wraps self.agent.ainvoke() with a retry loop. On failure, it waits self.base_delay * attempt seconds (linear backoff) before retrying, up to self.max_retries attempts. The default configuration uses max_retries=3 and base_delay=0.5 seconds.

Usage

Called internally by BaseAgent.run_trading_session() at each step of the reasoning loop. Not intended for direct external use.

Code Reference

Source Location

Repository: AI-Trader
File: agent/base_agent/base_agent.py
Lines: L423-435

Signature

async def _ainvoke_with_retry(self, message: List[Dict[str, str]]) -> Any:
    """
    Invoke the agent with retry logic.

    Args:
        message: Conversation messages to send to the agent

    Returns:
        Agent response dict with "messages" key

    Raises:
        Exception: Re-raises the last exception after max_retries exhausted
    """

Import

from agent.base_agent.base_agent import BaseAgent
# Internal method: agent._ainvoke_with_retry(message)

I/O Contract

Inputs

Name	Type	Required	Description
message	List[Dict[str, str]]	Yes	Conversation messages with role/content
self.max_retries	int	Config	Max retry attempts (default 3)
self.base_delay	float	Config	Base delay in seconds for backoff (default 0.5)

Outputs

Name	Type	Description
response	Any	Agent response dict from ainvoke() on success

Usage Examples

Internal Usage in Trading Loop

# Within BaseAgent.run_trading_session():
response = await self._ainvoke_with_retry(message)
agent_response = extract_conversation(response, "final")

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment