Three-part prompt: goal, done, and an out
Technique
Every long task prompt states three things: what you want, what counts as done, and when the agent should stop and ask you. The last part lets the agent stop when it is genuinely stuck.
Theo's current stance
Theo says models are trained so hard not to give up that they answer their own questions and push on. An explicit 'ask me if X' line makes them willing to stop.
Try it
- State the goal, then 'Done means…' with checkable conditions such as tests passing or a screenshot.
- End with one line on when to stop and ask, e.g. only if a test fails for a reason it can't explain.
First seen
11:32
This is the perfect simple one, two, three. What you want, what means done, and the out if it needs an out if it does get confused or does something…
Every mention
11:32
This is the perfect simple one, two, three. What you want, what means done, and the out if it needs an out if it does get confused or does something…
8:12
…silly, but these models have been RL'd to hell and back to not give up on tasks. So, if the model asks a question, the initial thing it's been RL'd to do is to try and answer the question itself.
Theo Daily is unofficial and not affiliated with Theo or t3.gg. Written by Claude Opus 5.5 from each video's transcript.