Context First: Prompts That Give AI What It Needs
Add the background, goal and constraints that turn a vague answer into a useful one.
A taste of a lesson
I asked it to write a project update for my boss and it was way too long and vague. What did I do wrong?
Probably nothing dramatic: the model just did not know enough. It could not see what your boss cares about, what changed this week, or how long the update should be, so it filled the gaps with generic padding. Try adding three lines: what happened (two or three facts), what you need from your boss, and the length, for example 'five bullet points, under 100 words'. Then compare both versions. Quick exercise: write the one sentence that says what your boss should do after reading it.
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Identify the four kinds of context a model cannot guess
- Rewrite a vague prompt into a specific one in a couple of minutes
- Judge when extra detail helps and when it only adds noise
- Recognise failures caused by missing facts rather than poor wording
Lesson plan
- 1 Why answers go generic Understand that a model fills missing details with the most typical answer it can produce. Start
- 2 Situation and goal Add the background facts and the real purpose behind a request. Start
- 3 Audience and constraints Describe the reader and the limits so the answer fits on the first try. Start
- 4 Placing and labelling reference material Paste documents and notes so the model can tell them apart from your instructions. Start
- 5 Too much, too little, conflicting Spot context that hurts the answer and trim it. Start
- 6 Your personal context checklist Build a short checklist you can apply to any request without thinking. Start
Try asking
About this tutor
For anyone who uses an AI assistant and keeps getting answers that are generic, too long or slightly off target. Most of those problems come from missing context, not a weak model. In these lessons you take your own real requests, such as an email, a plan or an explanation, and learn to add the four things a model cannot guess: the situation, the goal, the audience and the constraints. You rewrite prompts side by side, compare the results and build a short checklist you can use without thinking. By the end you will know how much context is enough, what to leave out, and how to tell when the problem is not the prompt at all.
Reviews
4.7
3 ratingsSample
- Marta K.Sample
The new colleague test is now stuck in my head. My weekly status emails take half the time because the first draft is close to right.
- Yuki T.Sample
I kept writing longer prompts thinking that was the fix. Learning to cut irrelevant detail made the biggest difference.
- Kwame A.Sample
Useful and short. I wanted a bit more on pasting long documents, but the lesson on conflicting context explained a problem I had for months.
About the teacher
Teaches prompting one technique at a time, with plenty of before and after examples
9 tutors 339 lessons taught Sample
I teach people to write prompts the way I once taught people to write briefs: say what you want, who it is for, and what good looks like. I came to AI from editing and corporate training, so I care more about clear thinking than clever tricks. In my lessons we take one technique at a time, try it on...
See Chiara's profile and tutorsMore like this
Other tutors on the same or nearby topics.