agent sidequestsIdeas worth trying
An ongoing helper

Send a bedtime idea. Wake up to a decision brief.

Let the passing idea become a small research job: a few grounded options, the real unknowns, and one sensible next move.

You get: A one-page decision briefOriginal prompt · Not run-tested

Your copy-and-paste prompt

Replace the [brackets] with your details, then paste this into your assistant.

Act as my small decision desk for this question: [question]. My constraints are [budget, time, location and preferences]. Spend no more than [research limit] on it and return a one-page brief by [date and time].

First confirm whether you can run in the background or schedule this work. If not, do the research now and say so. Find up to three plausible options, link the best primary sources, and separate facts, estimates and missing information. Recommend one reversible next step and explain what could change your recommendation. Record the question, options and conclusion in a document I can return to. Do not make purchases, submit forms or contact anyone. Stop when the brief is useful; a longer report is not the goal.
What the result could look like

The idea that survives until morning

A passing thought about learning pottery becomes three local routes, their published schedules, unanswered questions and a low-commitment first step.

Illustrative example. This is not an observed result or a verified recommendation.
The real-world spark

Inspired by a reported use

The author describes collecting daytime ideas and having an agent research them overnight before a morning review.

Andranik Sahakyan’s OpenClaw experiment · Jan 16, 2026

Firsthand self-report. We did not inspect the research runs or reproduce the workflow. The prompt above is our original extension, not the author's exact prompt or a tested result.

Make it more you

Try one question a week. Ask for updates only when new evidence changes a decision.

Get more from this idea

Keep the ideas coming

All sidequests