Hi:
I am trying to implement a playbook that uses an agent to summarize a conversation and then use that summary to update a provided care plan.
So the instructions are:
- summarize the conversation between the user and care manager using ${AGENT:Conversation summarizer}
- Use the summary from the above agent to update the care plan summary section only of the given care plan
- Output the updated care plan in the same format as the intput care plan
The user input I provide is as follows:
Conversation: 'Care Manager: Hey there! We all know the basics on what our heart does, but ' "have you ever wondered what's actually going on in that fist sized muscle we " 'User: To me it makes perfect sense.\n' 'Care Manager: Let me know if you have any questions or concerns ' 'after your appointment this afternoon!\n' 'User: Will do.. thank you\n'
Previous Care Plan: Care Plan Summary: The most important things for me to work on are... Maintaining a low fat diet rich in vegetables and high in essential vitamins and minerals in order to achieve a healthy cholesterol level daily Reducing my sodium intake and remaining compliant with medications in order to achieve a healthy blood pressure daily Following my physician's orders and attending all scheduled appointments Remaining physically active and compliant with prescribed medication regimen daily Take all medications as prescribed by your doctor. Do not adjust your medications without talking to your healthcare provider. Report any side effects of your medications to your healthcare provider. Keep track of your medication refills and make sure you don't run out. Medications may be prescribed to manage specific symptoms, such as pain, nausea, or anxiety.
But then the agent is not able to separate the conversation and the care plan. Keeps asking me to input the conversation separately. Attached is the conversation I had with the agent. How do I provide multiple inputs to be acted upon by different instructions?