Improve the suitability of the prompt that is sent to the OpenAI API. It should include permission for the LLM to admit it doesn't know an answer, and only give answers with a high degree of confidence. References to stocks need to be removed.
Acceptance criteria:
All stock references removed from prompt
Permission to answer "I don't know" included in prompt
Instruction to answer only with high confidence included in prompt
Improve the suitability of the prompt that is sent to the OpenAI API. It should include permission for the LLM to admit it doesn't know an answer, and only give answers with a high degree of confidence. References to stocks need to be removed.
Acceptance criteria: