LLMs have strict context length limits.

If system messages, instructions, and user input exceed this limit, earlier tokens are dropped silently. This often removes critical instructions.

Always calculate token usage explicitly and reserve space for the response. Truncate user input, not system prompts.

Common mistakes:

1. Assuming character count equals token count

2. Appending logs or history blindly

3. Ignoring model-specific context limits

Context budgeting is essential for reliable prompting.