📢 Announcing our research paper: Zentry achieves 26% higher accuracy than OpenAI Memory, 91% lower latency, and 90% token savings! Read the paper to learn how we're revolutionizing AI agent memory.

​
Introduction to Custom Instructions

Custom instructions allow you to define specific guidelines for your project. This feature helps ensure consistency and provides clear direction for handling project-specific requirements. Custom instructions are particularly useful when you want to:
  • Define how information should be extracted from conversations
  • Specify what types of data should be captured or ignored
  • Set rules for categorizing and organizing memories
  • Maintain consistent handling of project-specific requirements
When custom instructions are set at the project level, they will be applied to all new memories added within that project. This ensures that your data is processed according to your defined guidelines across your entire project.

​
Setting Custom Instructions

You can set custom instructions for your project using the following method: 
# Update custom instructions
prompt ="""
Your Task: Extract ONLY health-related information from conversations, focusing on the following areas:

1. Medical Conditions, Symptoms, and Diagnoses:
   - Illnesses, disorders, or symptoms (e.g., fever, diabetes).
   - Confirmed or suspected diagnoses.

2. Medications, Treatments, and Procedures:
   - Prescription or OTC medications (names, dosages).
   - Treatments, therapies, or medical procedures.

3. Diet, Exercise, and Sleep:
   - Dietary habits, fitness routines, and sleep patterns.

4. Doctor Visits and Appointments:
   - Past, upcoming, or regular medical visits.

5. Health Metrics:
   - Data like weight, BP, cholesterol, or sugar levels.

Guidelines:
- Focus solely on health-related content.
- Maintain clarity and context accuracy while recording.
"""
response = client.update_project(custom_instructions=prompt)
print(response)
You can also retrieve the current custom instructions: 
# Retrieve current custom instructions
response = client.get_project(fields=["custom_instructions"])
print(response)