[BUG]: Potential AI Response Handling Issue in Core Module

Describe the bug

During routine testing and integration with AIHawk core modules, inconsistencies were observed in AI response handling when processing edge cases with complex prompts and varied LLM configurations.

Steps to reproduce

1. Use branch named 'main'
2. Go to file containing AI response handler
3. Process a complex multi-turn conversation prompt
4. Execute with various LLM providers
5. Run program using command 'python -m aihawk'
6. Observe inconsistent response parsing behavior

Expected behavior

AI responses should be consistently parsed and handled regardless of complexity level and LLM provider, with proper error handling for edge cases.

Actual behavior

Occasional parsing failures and inconsistent handling of responses from different LLM providers, particularly with longer or more complex prompts.

Branch

main

Branch name

No response

Python version

3.10+

LLM Used

Multiple (ChatGPT, Claude, Gemini)

Model used

GPT-4, Claude-3, Gemini-Pro

Additional context

This issue impacts the reliability of AIHawk when used with different LLM providers and prompt complexities. A comprehensive review of response handling and parsing logic is recommended to ensure robustness across all supported configurations.