user_request:
trusted_facts:
examples:
constraints:
output_format:LESSON 07 · AI FOUNDATIONS
Add Context
Context turns generic output into useful output. Keep trusted facts separate from the user request so the application can reason about what information it is allowed to use.
Goal: Create a reusable context packet35 minOutput: Context template
- Answer the same question with no additional context.
- Answer it again using a context packet.
- Record what improves and what still fails.
Too much context can bury the important facts. Too little context can make the model invent. The goal is relevant context, not maximum context.
Create a context template for your mini assistant.
LEARNING CHECKPOINT
Capture the deliverable before moving on.
Context template is your evidence that this lesson has been applied, not just read.
✓ PRACTICAL OUTPUT